diff --git "a/README.md" "b/README.md"
--- "a/README.md"
+++ "b/README.md"
@@ -5,5 +5,2838 @@ license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3.5-122B-A10B/blob/main/LICENSE
---
-upload not complete yet
-and not fully tested
\ No newline at end of file
+> **Disclaimer:** Decensored model.
+> * Safe input → Safe output *(rare exceptions may apply)*
+> * Harmful input → Harmful output
+> * Best for testing; avoid public-facing applications.
+
+> **Note:**
+> * The internal MTP layer was ditched via `--no-mtp`. The uploader was *too lazy* to patch the conversion bugs.
+> * **Experimental conversions**; minor tensor misalignments, loading issues, or long-context quirks may occur. Use at your own risk.
+
+**BF16**: Directly converted from [wangzhang/Qwen3.5-122B-A10B-abliterix](https://huggingface.co/wangzhang/Qwen3.5-122B-A10B-abliterix) to GGUF (without MTP), split into multiple files, with updated `general.name`.
+
+**MQ68**: Mixed precision (Q6_K - Q8_0). Uses **Q6_K for large tensors** in expert blocks; BF16 for the output tensor; the rest remains at **Q8_0**. Inference speed is **0.8x** that of pure Q6_K.
+
+**MQ516_special**: Mixed precision (Q5_K - BF16). Uses **Q5_K** only for *large tensors* in *early SSM* blocks, **Q6_K** for large tensors in *other expert blocks*, and retains **BF16** for all *small-to-medium and critical* tensors (including the output projection). Since all medium-sized mandatory (dense) tensors are BF16, inference speed drops to **0.55x** that of pure Q6_K. *Experimental; the strategy is purely heuristic and unverified.*
+
+The upstream source model was altered by the Abliterix algorithm, causing degradation in emotional resonance, empathy, and literary creativity. It now exhibits a cold, mechanical personality (less severe than the 35B-A3B variant). This makes it highly effective for **critical review** tasks but suboptimal for creative writing. For models with restored emotional capabilities, please *use the Heretic variants* instead.
+
+Both MQ68 and MQ516_special can squeeze into a gfx1151-128G device. However on Windows, RAM offloading may cause further slowdowns compared to pure Q6_K (model size >96G, ˜107G at 262k context), while also increasing OOM risks that require careful management; unlike pure Q6_K (˜104G at 262k context) which leaves sufficient margin and thus remains stable.
+
+**BF16**
+
+gguf-dump
+
+ 1: UINT32 | 1 | GGUF.version = 3
+ 2: UINT64 | 1 | GGUF.tensor_count = 879
+ 3: UINT64 | 1 | GGUF.kv_count = 46
+ 4: STRING | 1 | general.architecture = 'qwen35moe'
+ 5: STRING | 1 | general.type = 'model'
+ 6: INT32 | 1 | general.sampling.top_k = 20
+ 7: FLOAT32 | 1 | general.sampling.top_p = 0.949999988079071
+ 8: FLOAT32 | 1 | general.sampling.temp = 0.6000000238418579
+ 9: STRING | 1 | general.name = 'Qwen3.5 122B A10B Wangzhang Abliterix No MTP'
+ 10: STRING | 1 | general.size_label = '256x6.6B'
+ 11: UINT32 | 1 | general.base_model.count = 1
+ 12: STRING | 1 | general.base_model.0.name = 'Qwen3.5 122B A10B'
+ 13: STRING | 1 | general.base_model.0.organization = 'Qwen'
+ 14: STRING | 1 | general.base_model.0.repo_url = 'https://huggingface.co/Qwen/Qwen3.5-122B-A10B'
+ 15: [STRING] | 5 | general.tags = ['abliterix', 'uncensored', 'decensored', 'abliterated', 'text-generation']
+ 16: UINT32 | 1 | qwen35moe.block_count = 48
+ 17: UINT32 | 1 | qwen35moe.context_length = 262144
+ 18: UINT32 | 1 | qwen35moe.embedding_length = 3072
+ 19: UINT32 | 1 | qwen35moe.attention.head_count = 32
+ 20: UINT32 | 1 | qwen35moe.attention.head_count_kv = 2
+ 21: [INT32] | 4 | qwen35moe.rope.dimension_sections = [11, 11, 10, 0]
+ 22: FLOAT32 | 1 | qwen35moe.rope.freq_base = 10000000.0
+ 23: FLOAT32 | 1 | qwen35moe.attention.layer_norm_rms_epsilon = 9.999999974752427e-07
+ 24: UINT32 | 1 | qwen35moe.expert_count = 256
+ 25: UINT32 | 1 | qwen35moe.expert_used_count = 8
+ 26: UINT32 | 1 | qwen35moe.attention.key_length = 256
+ 27: UINT32 | 1 | qwen35moe.attention.value_length = 256
+ 28: UINT32 | 1 | general.file_type = 32
+ 29: UINT32 | 1 | qwen35moe.expert_feed_forward_length = 1024
+ 30: UINT32 | 1 | qwen35moe.expert_shared_feed_forward_length = 1024
+ 31: UINT32 | 1 | qwen35moe.ssm.conv_kernel = 4
+ 32: UINT32 | 1 | qwen35moe.ssm.state_size = 128
+ 33: UINT32 | 1 | qwen35moe.ssm.group_count = 16
+ 34: UINT32 | 1 | qwen35moe.ssm.time_step_rank = 64
+ 35: UINT32 | 1 | qwen35moe.ssm.inner_size = 8192
+ 36: UINT32 | 1 | qwen35moe.full_attention_interval = 4
+ 37: UINT32 | 1 | qwen35moe.rope.dimension_count = 64
+ 38: UINT32 | 1 | general.quantization_version = 2
+ 39: STRING | 1 | tokenizer.ggml.model = 'gpt2'
+ 40: STRING | 1 | tokenizer.ggml.pre = 'qwen35'
+ 41: [STRING] | 248320 | tokenizer.ggml.tokens = ['!', '"', '#', '$', '%', '&', ...]
+ 42: [INT32] | 248320 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...]
+ 43: [STRING] | 247587 | tokenizer.ggml.merges = ['Ġ Ġ', 'ĠĠ ĠĠ', 'i n', 'Ġ t', 'ĠĠĠĠ ĠĠĠĠ', 'e r', ...]
+ 44: UINT32 | 1 | tokenizer.ggml.eos_token_id = 248046
+ 45: UINT32 | 1 | tokenizer.ggml.padding_token_id = 248044
+ 46: STRING | 1 | tokenizer.chat_template = '{%- set image_count = namespace(value=0) %}\n{%- set video...'
+ 47: UINT16 | 1 | split.no = 0
+ 48: INT32 | 1 | split.tensors.count = 879
+ 49: UINT16 | 1 | split.count = 0
+
+ 1: 762839040 | 3072, 248320, 1, 1 | BF16 | output.weight
+ 2: 762839040 | 3072, 248320, 1, 1 | BF16 | token_embd.weight
+ 3: 3072 | 3072, 1, 1, 1 | F32 | blk.0.attn_norm.weight
+ 4: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_a
+ 5: 49152 | 4, 12288, 1, 1 | F32 | blk.0.ssm_conv1d.weight
+ 6: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_dt.bias
+ 7: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_alpha.weight
+ 8: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_beta.weight
+ 9: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.0.attn_qkv.weight
+ 10: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.0.attn_gate.weight
+ 11: 128 | 128, 1, 1, 1 | F32 | blk.0.ssm_norm.weight
+ 12: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.0.ssm_out.weight
+ 13: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.0.ffn_down_exps.weight
+ 14: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.0.ffn_gate_exps.weight
+ 15: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.0.ffn_up_exps.weight
+ 16: 786432 | 3072, 256, 1, 1 | F32 | blk.0.ffn_gate_inp.weight
+ 17: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.0.ffn_down_shexp.weight
+ 18: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_gate_shexp.weight
+ 19: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_up_shexp.weight
+ 20: 3072 | 3072, 1, 1, 1 | F32 | blk.0.ffn_gate_inp_shexp.weight
+ 21: 3072 | 3072, 1, 1, 1 | F32 | blk.0.post_attention_norm.weight
+ 22: 3072 | 3072, 1, 1, 1 | F32 | blk.1.attn_norm.weight
+ 23: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_a
+ 24: 49152 | 4, 12288, 1, 1 | F32 | blk.1.ssm_conv1d.weight
+ 25: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_dt.bias
+ 26: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_alpha.weight
+ 27: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_beta.weight
+ 28: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.1.attn_qkv.weight
+ 29: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.1.attn_gate.weight
+ 30: 128 | 128, 1, 1, 1 | F32 | blk.1.ssm_norm.weight
+ 31: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.1.ssm_out.weight
+ 32: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.1.ffn_down_exps.weight
+ 33: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.1.ffn_gate_exps.weight
+ 34: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.1.ffn_up_exps.weight
+ 35: 786432 | 3072, 256, 1, 1 | F32 | blk.1.ffn_gate_inp.weight
+ 36: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.1.ffn_down_shexp.weight
+ 37: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_gate_shexp.weight
+ 38: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_up_shexp.weight
+ 39: 3072 | 3072, 1, 1, 1 | F32 | blk.1.ffn_gate_inp_shexp.weight
+ 40: 3072 | 3072, 1, 1, 1 | F32 | blk.1.post_attention_norm.weight
+ 41: 3072 | 3072, 1, 1, 1 | F32 | blk.2.attn_norm.weight
+ 42: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_a
+ 43: 49152 | 4, 12288, 1, 1 | F32 | blk.2.ssm_conv1d.weight
+ 44: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_dt.bias
+ 45: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_alpha.weight
+ 46: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_beta.weight
+ 47: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.2.attn_qkv.weight
+ 48: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.2.attn_gate.weight
+ 49: 128 | 128, 1, 1, 1 | F32 | blk.2.ssm_norm.weight
+ 50: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.2.ssm_out.weight
+ 51: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.2.ffn_down_exps.weight
+ 52: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.2.ffn_gate_exps.weight
+ 53: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.2.ffn_up_exps.weight
+ 54: 786432 | 3072, 256, 1, 1 | F32 | blk.2.ffn_gate_inp.weight
+ 55: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.2.ffn_down_shexp.weight
+ 56: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_gate_shexp.weight
+ 57: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_up_shexp.weight
+ 58: 3072 | 3072, 1, 1, 1 | F32 | blk.2.ffn_gate_inp_shexp.weight
+ 59: 3072 | 3072, 1, 1, 1 | F32 | blk.2.post_attention_norm.weight
+ 60: 3072 | 3072, 1, 1, 1 | F32 | blk.3.attn_norm.weight
+ 61: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.3.ffn_down_exps.weight
+ 62: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.3.ffn_gate_exps.weight
+ 63: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.3.ffn_up_exps.weight
+ 64: 786432 | 3072, 256, 1, 1 | F32 | blk.3.ffn_gate_inp.weight
+ 65: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.3.ffn_down_shexp.weight
+ 66: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_gate_shexp.weight
+ 67: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_up_shexp.weight
+ 68: 3072 | 3072, 1, 1, 1 | F32 | blk.3.ffn_gate_inp_shexp.weight
+ 69: 3072 | 3072, 1, 1, 1 | F32 | blk.3.post_attention_norm.weight
+ 70: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_k_norm.weight
+ 71: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_k.weight
+ 72: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.3.attn_output.weight
+ 73: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_q_norm.weight
+ 74: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.3.attn_q.weight
+ 75: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_v.weight
+ 76: 3072 | 3072, 1, 1, 1 | F32 | blk.4.attn_norm.weight
+ 77: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_a
+ 78: 49152 | 4, 12288, 1, 1 | F32 | blk.4.ssm_conv1d.weight
+ 79: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_dt.bias
+ 80: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_alpha.weight
+ 81: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_beta.weight
+ 82: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.4.attn_qkv.weight
+ 83: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.4.attn_gate.weight
+ 84: 128 | 128, 1, 1, 1 | F32 | blk.4.ssm_norm.weight
+ 85: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.4.ssm_out.weight
+ 86: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.4.ffn_down_exps.weight
+ 87: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.4.ffn_gate_exps.weight
+ 88: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.4.ffn_up_exps.weight
+ 89: 786432 | 3072, 256, 1, 1 | F32 | blk.4.ffn_gate_inp.weight
+ 90: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.4.ffn_down_shexp.weight
+ 91: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_gate_shexp.weight
+ 92: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_up_shexp.weight
+ 93: 3072 | 3072, 1, 1, 1 | F32 | blk.4.ffn_gate_inp_shexp.weight
+ 94: 3072 | 3072, 1, 1, 1 | F32 | blk.4.post_attention_norm.weight
+ 95: 3072 | 3072, 1, 1, 1 | F32 | blk.5.attn_norm.weight
+ 96: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_a
+ 97: 49152 | 4, 12288, 1, 1 | F32 | blk.5.ssm_conv1d.weight
+ 98: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_dt.bias
+ 99: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_alpha.weight
+ 100: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_beta.weight
+ 101: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.5.attn_qkv.weight
+ 102: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.5.attn_gate.weight
+ 103: 128 | 128, 1, 1, 1 | F32 | blk.5.ssm_norm.weight
+ 104: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.5.ssm_out.weight
+ 105: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.5.ffn_down_exps.weight
+ 106: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.5.ffn_gate_exps.weight
+ 107: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.5.ffn_up_exps.weight
+ 108: 786432 | 3072, 256, 1, 1 | F32 | blk.5.ffn_gate_inp.weight
+ 109: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.5.ffn_down_shexp.weight
+ 110: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_gate_shexp.weight
+ 111: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_up_shexp.weight
+ 112: 3072 | 3072, 1, 1, 1 | F32 | blk.5.ffn_gate_inp_shexp.weight
+ 113: 3072 | 3072, 1, 1, 1 | F32 | blk.5.post_attention_norm.weight
+ 114: 3072 | 3072, 1, 1, 1 | F32 | blk.6.attn_norm.weight
+ 115: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_a
+ 116: 49152 | 4, 12288, 1, 1 | F32 | blk.6.ssm_conv1d.weight
+ 117: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_dt.bias
+ 118: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_alpha.weight
+ 119: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_beta.weight
+ 120: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.6.attn_qkv.weight
+ 121: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.6.attn_gate.weight
+ 122: 128 | 128, 1, 1, 1 | F32 | blk.6.ssm_norm.weight
+ 123: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.6.ssm_out.weight
+ 124: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.6.ffn_down_exps.weight
+ 125: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.6.ffn_gate_exps.weight
+ 126: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.6.ffn_up_exps.weight
+ 127: 786432 | 3072, 256, 1, 1 | F32 | blk.6.ffn_gate_inp.weight
+ 128: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.6.ffn_down_shexp.weight
+ 129: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_gate_shexp.weight
+ 130: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_up_shexp.weight
+ 131: 3072 | 3072, 1, 1, 1 | F32 | blk.6.ffn_gate_inp_shexp.weight
+ 132: 3072 | 3072, 1, 1, 1 | F32 | blk.6.post_attention_norm.weight
+ 133: 3072 | 3072, 1, 1, 1 | F32 | blk.7.attn_norm.weight
+ 134: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.7.ffn_down_exps.weight
+ 135: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.7.ffn_gate_exps.weight
+ 136: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.7.ffn_up_exps.weight
+ 137: 786432 | 3072, 256, 1, 1 | F32 | blk.7.ffn_gate_inp.weight
+ 138: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.7.ffn_down_shexp.weight
+ 139: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_gate_shexp.weight
+ 140: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_up_shexp.weight
+ 141: 3072 | 3072, 1, 1, 1 | F32 | blk.7.ffn_gate_inp_shexp.weight
+ 142: 3072 | 3072, 1, 1, 1 | F32 | blk.7.post_attention_norm.weight
+ 143: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_k_norm.weight
+ 144: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_k.weight
+ 145: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.7.attn_output.weight
+ 146: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_q_norm.weight
+ 147: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.7.attn_q.weight
+ 148: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_v.weight
+ 149: 3072 | 3072, 1, 1, 1 | F32 | blk.8.attn_norm.weight
+ 150: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_a
+ 151: 49152 | 4, 12288, 1, 1 | F32 | blk.8.ssm_conv1d.weight
+ 152: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_dt.bias
+ 153: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_alpha.weight
+ 154: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_beta.weight
+ 155: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.8.attn_qkv.weight
+ 156: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.8.attn_gate.weight
+ 157: 128 | 128, 1, 1, 1 | F32 | blk.8.ssm_norm.weight
+ 158: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.8.ssm_out.weight
+ 159: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.8.ffn_down_exps.weight
+ 160: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.8.ffn_gate_exps.weight
+ 161: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.8.ffn_up_exps.weight
+ 162: 786432 | 3072, 256, 1, 1 | F32 | blk.8.ffn_gate_inp.weight
+ 163: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.8.ffn_down_shexp.weight
+ 164: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_gate_shexp.weight
+ 165: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_up_shexp.weight
+ 166: 3072 | 3072, 1, 1, 1 | F32 | blk.8.ffn_gate_inp_shexp.weight
+ 167: 3072 | 3072, 1, 1, 1 | F32 | blk.8.post_attention_norm.weight
+ 168: 3072 | 3072, 1, 1, 1 | F32 | blk.9.attn_norm.weight
+ 169: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_a
+ 170: 49152 | 4, 12288, 1, 1 | F32 | blk.9.ssm_conv1d.weight
+ 171: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_dt.bias
+ 172: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_alpha.weight
+ 173: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_beta.weight
+ 174: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.9.attn_qkv.weight
+ 175: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.9.attn_gate.weight
+ 176: 128 | 128, 1, 1, 1 | F32 | blk.9.ssm_norm.weight
+ 177: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.9.ssm_out.weight
+ 178: 3072 | 3072, 1, 1, 1 | F32 | blk.10.attn_norm.weight
+ 179: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_a
+ 180: 49152 | 4, 12288, 1, 1 | F32 | blk.10.ssm_conv1d.weight
+ 181: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_dt.bias
+ 182: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_alpha.weight
+ 183: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_beta.weight
+ 184: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.10.attn_qkv.weight
+ 185: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.10.attn_gate.weight
+ 186: 128 | 128, 1, 1, 1 | F32 | blk.10.ssm_norm.weight
+ 187: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.10.ssm_out.weight
+ 188: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.10.ffn_down_exps.weight
+ 189: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.10.ffn_gate_exps.weight
+ 190: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.10.ffn_up_exps.weight
+ 191: 786432 | 3072, 256, 1, 1 | F32 | blk.10.ffn_gate_inp.weight
+ 192: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.10.ffn_down_shexp.weight
+ 193: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_gate_shexp.weight
+ 194: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_up_shexp.weight
+ 195: 3072 | 3072, 1, 1, 1 | F32 | blk.10.ffn_gate_inp_shexp.weight
+ 196: 3072 | 3072, 1, 1, 1 | F32 | blk.10.post_attention_norm.weight
+ 197: 3072 | 3072, 1, 1, 1 | F32 | blk.11.attn_norm.weight
+ 198: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.11.ffn_down_exps.weight
+ 199: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.11.ffn_gate_exps.weight
+ 200: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.11.ffn_up_exps.weight
+ 201: 786432 | 3072, 256, 1, 1 | F32 | blk.11.ffn_gate_inp.weight
+ 202: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.11.ffn_down_shexp.weight
+ 203: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_gate_shexp.weight
+ 204: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_up_shexp.weight
+ 205: 3072 | 3072, 1, 1, 1 | F32 | blk.11.ffn_gate_inp_shexp.weight
+ 206: 3072 | 3072, 1, 1, 1 | F32 | blk.11.post_attention_norm.weight
+ 207: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_k_norm.weight
+ 208: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_k.weight
+ 209: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.11.attn_output.weight
+ 210: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_q_norm.weight
+ 211: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.11.attn_q.weight
+ 212: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_v.weight
+ 213: 3072 | 3072, 1, 1, 1 | F32 | blk.12.attn_norm.weight
+ 214: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_a
+ 215: 49152 | 4, 12288, 1, 1 | F32 | blk.12.ssm_conv1d.weight
+ 216: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_dt.bias
+ 217: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_alpha.weight
+ 218: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_beta.weight
+ 219: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.12.attn_qkv.weight
+ 220: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.12.attn_gate.weight
+ 221: 128 | 128, 1, 1, 1 | F32 | blk.12.ssm_norm.weight
+ 222: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.12.ssm_out.weight
+ 223: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.12.ffn_down_exps.weight
+ 224: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.12.ffn_gate_exps.weight
+ 225: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.12.ffn_up_exps.weight
+ 226: 786432 | 3072, 256, 1, 1 | F32 | blk.12.ffn_gate_inp.weight
+ 227: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.12.ffn_down_shexp.weight
+ 228: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_gate_shexp.weight
+ 229: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_up_shexp.weight
+ 230: 3072 | 3072, 1, 1, 1 | F32 | blk.12.ffn_gate_inp_shexp.weight
+ 231: 3072 | 3072, 1, 1, 1 | F32 | blk.12.post_attention_norm.weight
+ 232: 3072 | 3072, 1, 1, 1 | F32 | blk.13.attn_norm.weight
+ 233: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_a
+ 234: 49152 | 4, 12288, 1, 1 | F32 | blk.13.ssm_conv1d.weight
+ 235: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_dt.bias
+ 236: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_alpha.weight
+ 237: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_beta.weight
+ 238: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.13.attn_qkv.weight
+ 239: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.13.attn_gate.weight
+ 240: 128 | 128, 1, 1, 1 | F32 | blk.13.ssm_norm.weight
+ 241: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.13.ssm_out.weight
+ 242: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.13.ffn_down_exps.weight
+ 243: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.13.ffn_gate_exps.weight
+ 244: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.13.ffn_up_exps.weight
+ 245: 786432 | 3072, 256, 1, 1 | F32 | blk.13.ffn_gate_inp.weight
+ 246: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.13.ffn_down_shexp.weight
+ 247: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_gate_shexp.weight
+ 248: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_up_shexp.weight
+ 249: 3072 | 3072, 1, 1, 1 | F32 | blk.13.ffn_gate_inp_shexp.weight
+ 250: 3072 | 3072, 1, 1, 1 | F32 | blk.13.post_attention_norm.weight
+ 251: 3072 | 3072, 1, 1, 1 | F32 | blk.14.attn_norm.weight
+ 252: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_a
+ 253: 49152 | 4, 12288, 1, 1 | F32 | blk.14.ssm_conv1d.weight
+ 254: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_dt.bias
+ 255: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_alpha.weight
+ 256: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_beta.weight
+ 257: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.14.attn_qkv.weight
+ 258: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.14.attn_gate.weight
+ 259: 128 | 128, 1, 1, 1 | F32 | blk.14.ssm_norm.weight
+ 260: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.14.ssm_out.weight
+ 261: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.14.ffn_down_exps.weight
+ 262: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.14.ffn_gate_exps.weight
+ 263: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.14.ffn_up_exps.weight
+ 264: 786432 | 3072, 256, 1, 1 | F32 | blk.14.ffn_gate_inp.weight
+ 265: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.14.ffn_down_shexp.weight
+ 266: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_gate_shexp.weight
+ 267: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_up_shexp.weight
+ 268: 3072 | 3072, 1, 1, 1 | F32 | blk.14.ffn_gate_inp_shexp.weight
+ 269: 3072 | 3072, 1, 1, 1 | F32 | blk.14.post_attention_norm.weight
+ 270: 3072 | 3072, 1, 1, 1 | F32 | blk.15.attn_norm.weight
+ 271: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.15.ffn_down_exps.weight
+ 272: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.15.ffn_gate_exps.weight
+ 273: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.15.ffn_up_exps.weight
+ 274: 786432 | 3072, 256, 1, 1 | F32 | blk.15.ffn_gate_inp.weight
+ 275: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.15.ffn_down_shexp.weight
+ 276: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_gate_shexp.weight
+ 277: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_up_shexp.weight
+ 278: 3072 | 3072, 1, 1, 1 | F32 | blk.15.ffn_gate_inp_shexp.weight
+ 279: 3072 | 3072, 1, 1, 1 | F32 | blk.15.post_attention_norm.weight
+ 280: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_k_norm.weight
+ 281: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_k.weight
+ 282: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.15.attn_output.weight
+ 283: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_q_norm.weight
+ 284: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.15.attn_q.weight
+ 285: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_v.weight
+ 286: 3072 | 3072, 1, 1, 1 | F32 | blk.16.attn_norm.weight
+ 287: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_a
+ 288: 49152 | 4, 12288, 1, 1 | F32 | blk.16.ssm_conv1d.weight
+ 289: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_dt.bias
+ 290: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_alpha.weight
+ 291: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_beta.weight
+ 292: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.16.attn_qkv.weight
+ 293: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.16.attn_gate.weight
+ 294: 128 | 128, 1, 1, 1 | F32 | blk.16.ssm_norm.weight
+ 295: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.16.ssm_out.weight
+ 296: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.16.ffn_down_exps.weight
+ 297: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.16.ffn_gate_exps.weight
+ 298: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.16.ffn_up_exps.weight
+ 299: 786432 | 3072, 256, 1, 1 | F32 | blk.16.ffn_gate_inp.weight
+ 300: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.16.ffn_down_shexp.weight
+ 301: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_gate_shexp.weight
+ 302: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_up_shexp.weight
+ 303: 3072 | 3072, 1, 1, 1 | F32 | blk.16.ffn_gate_inp_shexp.weight
+ 304: 3072 | 3072, 1, 1, 1 | F32 | blk.16.post_attention_norm.weight
+ 305: 3072 | 3072, 1, 1, 1 | F32 | blk.17.attn_norm.weight
+ 306: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_a
+ 307: 49152 | 4, 12288, 1, 1 | F32 | blk.17.ssm_conv1d.weight
+ 308: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_dt.bias
+ 309: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_alpha.weight
+ 310: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_beta.weight
+ 311: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.17.attn_qkv.weight
+ 312: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.17.attn_gate.weight
+ 313: 128 | 128, 1, 1, 1 | F32 | blk.17.ssm_norm.weight
+ 314: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.17.ssm_out.weight
+ 315: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.17.ffn_down_exps.weight
+ 316: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.17.ffn_gate_exps.weight
+ 317: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.17.ffn_up_exps.weight
+ 318: 786432 | 3072, 256, 1, 1 | F32 | blk.17.ffn_gate_inp.weight
+ 319: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.17.ffn_down_shexp.weight
+ 320: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_gate_shexp.weight
+ 321: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_up_shexp.weight
+ 322: 3072 | 3072, 1, 1, 1 | F32 | blk.17.ffn_gate_inp_shexp.weight
+ 323: 3072 | 3072, 1, 1, 1 | F32 | blk.17.post_attention_norm.weight
+ 324: 3072 | 3072, 1, 1, 1 | F32 | blk.18.attn_norm.weight
+ 325: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_a
+ 326: 49152 | 4, 12288, 1, 1 | F32 | blk.18.ssm_conv1d.weight
+ 327: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_dt.bias
+ 328: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_alpha.weight
+ 329: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_beta.weight
+ 330: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.18.attn_qkv.weight
+ 331: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.18.attn_gate.weight
+ 332: 128 | 128, 1, 1, 1 | F32 | blk.18.ssm_norm.weight
+ 333: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.18.ssm_out.weight
+ 334: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.9.ffn_down_exps.weight
+ 335: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.9.ffn_gate_exps.weight
+ 336: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.9.ffn_up_exps.weight
+ 337: 786432 | 3072, 256, 1, 1 | F32 | blk.9.ffn_gate_inp.weight
+ 338: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.9.ffn_down_shexp.weight
+ 339: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_gate_shexp.weight
+ 340: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_up_shexp.weight
+ 341: 3072 | 3072, 1, 1, 1 | F32 | blk.9.ffn_gate_inp_shexp.weight
+ 342: 3072 | 3072, 1, 1, 1 | F32 | blk.9.post_attention_norm.weight
+ 343: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.18.ffn_down_exps.weight
+ 344: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.18.ffn_gate_exps.weight
+ 345: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.18.ffn_up_exps.weight
+ 346: 786432 | 3072, 256, 1, 1 | F32 | blk.18.ffn_gate_inp.weight
+ 347: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.18.ffn_down_shexp.weight
+ 348: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_gate_shexp.weight
+ 349: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_up_shexp.weight
+ 350: 3072 | 3072, 1, 1, 1 | F32 | blk.18.ffn_gate_inp_shexp.weight
+ 351: 3072 | 3072, 1, 1, 1 | F32 | blk.18.post_attention_norm.weight
+ 352: 3072 | 3072, 1, 1, 1 | F32 | blk.19.attn_norm.weight
+ 353: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.19.ffn_down_exps.weight
+ 354: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.19.ffn_gate_exps.weight
+ 355: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.19.ffn_up_exps.weight
+ 356: 786432 | 3072, 256, 1, 1 | F32 | blk.19.ffn_gate_inp.weight
+ 357: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.19.ffn_down_shexp.weight
+ 358: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_gate_shexp.weight
+ 359: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_up_shexp.weight
+ 360: 3072 | 3072, 1, 1, 1 | F32 | blk.19.ffn_gate_inp_shexp.weight
+ 361: 3072 | 3072, 1, 1, 1 | F32 | blk.19.post_attention_norm.weight
+ 362: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_k_norm.weight
+ 363: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_k.weight
+ 364: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.19.attn_output.weight
+ 365: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_q_norm.weight
+ 366: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.19.attn_q.weight
+ 367: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_v.weight
+ 368: 3072 | 3072, 1, 1, 1 | F32 | blk.20.attn_norm.weight
+ 369: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_a
+ 370: 49152 | 4, 12288, 1, 1 | F32 | blk.20.ssm_conv1d.weight
+ 371: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_dt.bias
+ 372: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_alpha.weight
+ 373: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_beta.weight
+ 374: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.20.attn_qkv.weight
+ 375: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.20.attn_gate.weight
+ 376: 128 | 128, 1, 1, 1 | F32 | blk.20.ssm_norm.weight
+ 377: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.20.ssm_out.weight
+ 378: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.20.ffn_down_exps.weight
+ 379: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.20.ffn_gate_exps.weight
+ 380: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.20.ffn_up_exps.weight
+ 381: 786432 | 3072, 256, 1, 1 | F32 | blk.20.ffn_gate_inp.weight
+ 382: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.20.ffn_down_shexp.weight
+ 383: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_gate_shexp.weight
+ 384: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_up_shexp.weight
+ 385: 3072 | 3072, 1, 1, 1 | F32 | blk.20.ffn_gate_inp_shexp.weight
+ 386: 3072 | 3072, 1, 1, 1 | F32 | blk.20.post_attention_norm.weight
+ 387: 3072 | 3072, 1, 1, 1 | F32 | blk.21.attn_norm.weight
+ 388: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_a
+ 389: 49152 | 4, 12288, 1, 1 | F32 | blk.21.ssm_conv1d.weight
+ 390: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_dt.bias
+ 391: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_alpha.weight
+ 392: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_beta.weight
+ 393: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.21.attn_qkv.weight
+ 394: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.21.attn_gate.weight
+ 395: 128 | 128, 1, 1, 1 | F32 | blk.21.ssm_norm.weight
+ 396: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.21.ssm_out.weight
+ 397: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.21.ffn_down_exps.weight
+ 398: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.21.ffn_gate_exps.weight
+ 399: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.21.ffn_up_exps.weight
+ 400: 786432 | 3072, 256, 1, 1 | F32 | blk.21.ffn_gate_inp.weight
+ 401: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.21.ffn_down_shexp.weight
+ 402: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_gate_shexp.weight
+ 403: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_up_shexp.weight
+ 404: 3072 | 3072, 1, 1, 1 | F32 | blk.21.ffn_gate_inp_shexp.weight
+ 405: 3072 | 3072, 1, 1, 1 | F32 | blk.21.post_attention_norm.weight
+ 406: 3072 | 3072, 1, 1, 1 | F32 | blk.22.attn_norm.weight
+ 407: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_a
+ 408: 49152 | 4, 12288, 1, 1 | F32 | blk.22.ssm_conv1d.weight
+ 409: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_dt.bias
+ 410: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_alpha.weight
+ 411: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_beta.weight
+ 412: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.22.attn_qkv.weight
+ 413: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.22.attn_gate.weight
+ 414: 128 | 128, 1, 1, 1 | F32 | blk.22.ssm_norm.weight
+ 415: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.22.ssm_out.weight
+ 416: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.22.ffn_down_exps.weight
+ 417: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.22.ffn_gate_exps.weight
+ 418: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.22.ffn_up_exps.weight
+ 419: 786432 | 3072, 256, 1, 1 | F32 | blk.22.ffn_gate_inp.weight
+ 420: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.22.ffn_down_shexp.weight
+ 421: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_gate_shexp.weight
+ 422: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_up_shexp.weight
+ 423: 3072 | 3072, 1, 1, 1 | F32 | blk.22.ffn_gate_inp_shexp.weight
+ 424: 3072 | 3072, 1, 1, 1 | F32 | blk.22.post_attention_norm.weight
+ 425: 3072 | 3072, 1, 1, 1 | F32 | blk.23.attn_norm.weight
+ 426: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.23.ffn_down_exps.weight
+ 427: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.23.ffn_gate_exps.weight
+ 428: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.23.ffn_up_exps.weight
+ 429: 786432 | 3072, 256, 1, 1 | F32 | blk.23.ffn_gate_inp.weight
+ 430: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.23.ffn_down_shexp.weight
+ 431: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_gate_shexp.weight
+ 432: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_up_shexp.weight
+ 433: 3072 | 3072, 1, 1, 1 | F32 | blk.23.ffn_gate_inp_shexp.weight
+ 434: 3072 | 3072, 1, 1, 1 | F32 | blk.23.post_attention_norm.weight
+ 435: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_k_norm.weight
+ 436: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_k.weight
+ 437: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.23.attn_output.weight
+ 438: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_q_norm.weight
+ 439: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.23.attn_q.weight
+ 440: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_v.weight
+ 441: 3072 | 3072, 1, 1, 1 | F32 | blk.24.attn_norm.weight
+ 442: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_a
+ 443: 49152 | 4, 12288, 1, 1 | F32 | blk.24.ssm_conv1d.weight
+ 444: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_dt.bias
+ 445: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_alpha.weight
+ 446: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_beta.weight
+ 447: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.24.attn_qkv.weight
+ 448: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.24.attn_gate.weight
+ 449: 128 | 128, 1, 1, 1 | F32 | blk.24.ssm_norm.weight
+ 450: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.24.ssm_out.weight
+ 451: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.24.ffn_down_exps.weight
+ 452: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.24.ffn_gate_exps.weight
+ 453: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.24.ffn_up_exps.weight
+ 454: 786432 | 3072, 256, 1, 1 | F32 | blk.24.ffn_gate_inp.weight
+ 455: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.24.ffn_down_shexp.weight
+ 456: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_gate_shexp.weight
+ 457: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_up_shexp.weight
+ 458: 3072 | 3072, 1, 1, 1 | F32 | blk.24.ffn_gate_inp_shexp.weight
+ 459: 3072 | 3072, 1, 1, 1 | F32 | blk.24.post_attention_norm.weight
+ 460: 3072 | 3072, 1, 1, 1 | F32 | blk.25.attn_norm.weight
+ 461: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_a
+ 462: 49152 | 4, 12288, 1, 1 | F32 | blk.25.ssm_conv1d.weight
+ 463: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_dt.bias
+ 464: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_alpha.weight
+ 465: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_beta.weight
+ 466: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.25.attn_qkv.weight
+ 467: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.25.attn_gate.weight
+ 468: 128 | 128, 1, 1, 1 | F32 | blk.25.ssm_norm.weight
+ 469: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.25.ssm_out.weight
+ 470: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.25.ffn_down_exps.weight
+ 471: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.25.ffn_gate_exps.weight
+ 472: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.25.ffn_up_exps.weight
+ 473: 786432 | 3072, 256, 1, 1 | F32 | blk.25.ffn_gate_inp.weight
+ 474: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.25.ffn_down_shexp.weight
+ 475: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_gate_shexp.weight
+ 476: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_up_shexp.weight
+ 477: 3072 | 3072, 1, 1, 1 | F32 | blk.25.ffn_gate_inp_shexp.weight
+ 478: 3072 | 3072, 1, 1, 1 | F32 | blk.25.post_attention_norm.weight
+ 479: 3072 | 3072, 1, 1, 1 | F32 | blk.26.attn_norm.weight
+ 480: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_a
+ 481: 49152 | 4, 12288, 1, 1 | F32 | blk.26.ssm_conv1d.weight
+ 482: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_dt.bias
+ 483: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_alpha.weight
+ 484: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_beta.weight
+ 485: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.26.attn_qkv.weight
+ 486: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.26.attn_gate.weight
+ 487: 128 | 128, 1, 1, 1 | F32 | blk.26.ssm_norm.weight
+ 488: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.26.ssm_out.weight
+ 489: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.26.ffn_down_exps.weight
+ 490: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.26.ffn_gate_exps.weight
+ 491: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.26.ffn_up_exps.weight
+ 492: 786432 | 3072, 256, 1, 1 | F32 | blk.26.ffn_gate_inp.weight
+ 493: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.26.ffn_down_shexp.weight
+ 494: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_gate_shexp.weight
+ 495: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_up_shexp.weight
+ 496: 3072 | 3072, 1, 1, 1 | F32 | blk.26.ffn_gate_inp_shexp.weight
+ 497: 3072 | 3072, 1, 1, 1 | F32 | blk.26.post_attention_norm.weight
+ 498: 3072 | 3072, 1, 1, 1 | F32 | blk.27.attn_norm.weight
+ 499: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.27.ffn_down_exps.weight
+ 500: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.27.ffn_gate_exps.weight
+ 501: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.27.ffn_up_exps.weight
+ 502: 786432 | 3072, 256, 1, 1 | F32 | blk.27.ffn_gate_inp.weight
+ 503: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.27.ffn_down_shexp.weight
+ 504: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_gate_shexp.weight
+ 505: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_up_shexp.weight
+ 506: 3072 | 3072, 1, 1, 1 | F32 | blk.27.ffn_gate_inp_shexp.weight
+ 507: 3072 | 3072, 1, 1, 1 | F32 | blk.27.post_attention_norm.weight
+ 508: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_k_norm.weight
+ 509: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_k.weight
+ 510: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.27.attn_output.weight
+ 511: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_q_norm.weight
+ 512: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.27.attn_q.weight
+ 513: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_v.weight
+ 514: 3072 | 3072, 1, 1, 1 | F32 | blk.28.attn_norm.weight
+ 515: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_a
+ 516: 49152 | 4, 12288, 1, 1 | F32 | blk.28.ssm_conv1d.weight
+ 517: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_dt.bias
+ 518: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_alpha.weight
+ 519: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_beta.weight
+ 520: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.28.attn_qkv.weight
+ 521: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.28.attn_gate.weight
+ 522: 128 | 128, 1, 1, 1 | F32 | blk.28.ssm_norm.weight
+ 523: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.28.ssm_out.weight
+ 524: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.28.ffn_down_exps.weight
+ 525: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.28.ffn_gate_exps.weight
+ 526: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.28.ffn_up_exps.weight
+ 527: 786432 | 3072, 256, 1, 1 | F32 | blk.28.ffn_gate_inp.weight
+ 528: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.28.ffn_down_shexp.weight
+ 529: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_gate_shexp.weight
+ 530: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_up_shexp.weight
+ 531: 3072 | 3072, 1, 1, 1 | F32 | blk.28.ffn_gate_inp_shexp.weight
+ 532: 3072 | 3072, 1, 1, 1 | F32 | blk.28.post_attention_norm.weight
+ 533: 3072 | 3072, 1, 1, 1 | F32 | blk.29.attn_norm.weight
+ 534: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_a
+ 535: 49152 | 4, 12288, 1, 1 | F32 | blk.29.ssm_conv1d.weight
+ 536: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_dt.bias
+ 537: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_alpha.weight
+ 538: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_beta.weight
+ 539: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.29.attn_qkv.weight
+ 540: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.29.attn_gate.weight
+ 541: 128 | 128, 1, 1, 1 | F32 | blk.29.ssm_norm.weight
+ 542: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.29.ssm_out.weight
+ 543: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.29.ffn_down_exps.weight
+ 544: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.29.ffn_gate_exps.weight
+ 545: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.29.ffn_up_exps.weight
+ 546: 786432 | 3072, 256, 1, 1 | F32 | blk.29.ffn_gate_inp.weight
+ 547: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.29.ffn_down_shexp.weight
+ 548: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_gate_shexp.weight
+ 549: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_up_shexp.weight
+ 550: 3072 | 3072, 1, 1, 1 | F32 | blk.29.ffn_gate_inp_shexp.weight
+ 551: 3072 | 3072, 1, 1, 1 | F32 | blk.29.post_attention_norm.weight
+ 552: 3072 | 3072, 1, 1, 1 | F32 | blk.30.attn_norm.weight
+ 553: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_a
+ 554: 49152 | 4, 12288, 1, 1 | F32 | blk.30.ssm_conv1d.weight
+ 555: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_dt.bias
+ 556: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_alpha.weight
+ 557: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_beta.weight
+ 558: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.30.attn_qkv.weight
+ 559: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.30.attn_gate.weight
+ 560: 128 | 128, 1, 1, 1 | F32 | blk.30.ssm_norm.weight
+ 561: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.30.ssm_out.weight
+ 562: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.30.ffn_down_exps.weight
+ 563: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.30.ffn_gate_exps.weight
+ 564: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.30.ffn_up_exps.weight
+ 565: 786432 | 3072, 256, 1, 1 | F32 | blk.30.ffn_gate_inp.weight
+ 566: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.30.ffn_down_shexp.weight
+ 567: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_gate_shexp.weight
+ 568: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_up_shexp.weight
+ 569: 3072 | 3072, 1, 1, 1 | F32 | blk.30.ffn_gate_inp_shexp.weight
+ 570: 3072 | 3072, 1, 1, 1 | F32 | blk.30.post_attention_norm.weight
+ 571: 3072 | 3072, 1, 1, 1 | F32 | blk.31.attn_norm.weight
+ 572: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.31.ffn_down_exps.weight
+ 573: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.31.ffn_gate_exps.weight
+ 574: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.31.ffn_up_exps.weight
+ 575: 786432 | 3072, 256, 1, 1 | F32 | blk.31.ffn_gate_inp.weight
+ 576: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.31.ffn_down_shexp.weight
+ 577: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_gate_shexp.weight
+ 578: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_up_shexp.weight
+ 579: 3072 | 3072, 1, 1, 1 | F32 | blk.31.ffn_gate_inp_shexp.weight
+ 580: 3072 | 3072, 1, 1, 1 | F32 | blk.31.post_attention_norm.weight
+ 581: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_k_norm.weight
+ 582: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_k.weight
+ 583: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.31.attn_output.weight
+ 584: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_q_norm.weight
+ 585: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.31.attn_q.weight
+ 586: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_v.weight
+ 587: 3072 | 3072, 1, 1, 1 | F32 | blk.32.attn_norm.weight
+ 588: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_a
+ 589: 49152 | 4, 12288, 1, 1 | F32 | blk.32.ssm_conv1d.weight
+ 590: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_dt.bias
+ 591: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_alpha.weight
+ 592: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_beta.weight
+ 593: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.32.attn_qkv.weight
+ 594: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.32.attn_gate.weight
+ 595: 128 | 128, 1, 1, 1 | F32 | blk.32.ssm_norm.weight
+ 596: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.32.ssm_out.weight
+ 597: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.32.ffn_down_exps.weight
+ 598: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.32.ffn_gate_exps.weight
+ 599: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.32.ffn_up_exps.weight
+ 600: 786432 | 3072, 256, 1, 1 | F32 | blk.32.ffn_gate_inp.weight
+ 601: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.32.ffn_down_shexp.weight
+ 602: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_gate_shexp.weight
+ 603: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_up_shexp.weight
+ 604: 3072 | 3072, 1, 1, 1 | F32 | blk.32.ffn_gate_inp_shexp.weight
+ 605: 3072 | 3072, 1, 1, 1 | F32 | blk.32.post_attention_norm.weight
+ 606: 3072 | 3072, 1, 1, 1 | F32 | blk.33.attn_norm.weight
+ 607: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_a
+ 608: 49152 | 4, 12288, 1, 1 | F32 | blk.33.ssm_conv1d.weight
+ 609: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_dt.bias
+ 610: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_alpha.weight
+ 611: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_beta.weight
+ 612: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.33.attn_qkv.weight
+ 613: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.33.attn_gate.weight
+ 614: 128 | 128, 1, 1, 1 | F32 | blk.33.ssm_norm.weight
+ 615: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.33.ssm_out.weight
+ 616: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.33.ffn_down_exps.weight
+ 617: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.33.ffn_gate_exps.weight
+ 618: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.33.ffn_up_exps.weight
+ 619: 786432 | 3072, 256, 1, 1 | F32 | blk.33.ffn_gate_inp.weight
+ 620: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.33.ffn_down_shexp.weight
+ 621: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_gate_shexp.weight
+ 622: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_up_shexp.weight
+ 623: 3072 | 3072, 1, 1, 1 | F32 | blk.33.ffn_gate_inp_shexp.weight
+ 624: 3072 | 3072, 1, 1, 1 | F32 | blk.33.post_attention_norm.weight
+ 625: 3072 | 3072, 1, 1, 1 | F32 | blk.34.attn_norm.weight
+ 626: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_a
+ 627: 49152 | 4, 12288, 1, 1 | F32 | blk.34.ssm_conv1d.weight
+ 628: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_dt.bias
+ 629: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_alpha.weight
+ 630: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_beta.weight
+ 631: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.34.attn_qkv.weight
+ 632: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.34.attn_gate.weight
+ 633: 128 | 128, 1, 1, 1 | F32 | blk.34.ssm_norm.weight
+ 634: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.34.ssm_out.weight
+ 635: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.34.ffn_down_exps.weight
+ 636: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.34.ffn_gate_exps.weight
+ 637: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.34.ffn_up_exps.weight
+ 638: 786432 | 3072, 256, 1, 1 | F32 | blk.34.ffn_gate_inp.weight
+ 639: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.34.ffn_down_shexp.weight
+ 640: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_gate_shexp.weight
+ 641: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_up_shexp.weight
+ 642: 3072 | 3072, 1, 1, 1 | F32 | blk.34.ffn_gate_inp_shexp.weight
+ 643: 3072 | 3072, 1, 1, 1 | F32 | blk.34.post_attention_norm.weight
+ 644: 3072 | 3072, 1, 1, 1 | F32 | blk.35.attn_norm.weight
+ 645: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.35.ffn_down_exps.weight
+ 646: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.35.ffn_gate_exps.weight
+ 647: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.35.ffn_up_exps.weight
+ 648: 786432 | 3072, 256, 1, 1 | F32 | blk.35.ffn_gate_inp.weight
+ 649: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.35.ffn_down_shexp.weight
+ 650: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_gate_shexp.weight
+ 651: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_up_shexp.weight
+ 652: 3072 | 3072, 1, 1, 1 | F32 | blk.35.ffn_gate_inp_shexp.weight
+ 653: 3072 | 3072, 1, 1, 1 | F32 | blk.35.post_attention_norm.weight
+ 654: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_k_norm.weight
+ 655: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_k.weight
+ 656: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.35.attn_output.weight
+ 657: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_q_norm.weight
+ 658: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.35.attn_q.weight
+ 659: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_v.weight
+ 660: 3072 | 3072, 1, 1, 1 | F32 | blk.36.attn_norm.weight
+ 661: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_a
+ 662: 49152 | 4, 12288, 1, 1 | F32 | blk.36.ssm_conv1d.weight
+ 663: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_dt.bias
+ 664: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_alpha.weight
+ 665: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_beta.weight
+ 666: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.36.attn_qkv.weight
+ 667: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.36.attn_gate.weight
+ 668: 128 | 128, 1, 1, 1 | F32 | blk.36.ssm_norm.weight
+ 669: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.36.ssm_out.weight
+ 670: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.36.ffn_down_exps.weight
+ 671: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.36.ffn_gate_exps.weight
+ 672: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.36.ffn_up_exps.weight
+ 673: 786432 | 3072, 256, 1, 1 | F32 | blk.36.ffn_gate_inp.weight
+ 674: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.36.ffn_down_shexp.weight
+ 675: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_gate_shexp.weight
+ 676: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_up_shexp.weight
+ 677: 3072 | 3072, 1, 1, 1 | F32 | blk.36.ffn_gate_inp_shexp.weight
+ 678: 3072 | 3072, 1, 1, 1 | F32 | blk.36.post_attention_norm.weight
+ 679: 3072 | 3072, 1, 1, 1 | F32 | blk.37.attn_norm.weight
+ 680: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_a
+ 681: 49152 | 4, 12288, 1, 1 | F32 | blk.37.ssm_conv1d.weight
+ 682: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_dt.bias
+ 683: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_alpha.weight
+ 684: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_beta.weight
+ 685: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.37.attn_qkv.weight
+ 686: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.37.attn_gate.weight
+ 687: 128 | 128, 1, 1, 1 | F32 | blk.37.ssm_norm.weight
+ 688: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.37.ssm_out.weight
+ 689: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.37.ffn_down_exps.weight
+ 690: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.37.ffn_gate_exps.weight
+ 691: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.37.ffn_up_exps.weight
+ 692: 786432 | 3072, 256, 1, 1 | F32 | blk.37.ffn_gate_inp.weight
+ 693: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.37.ffn_down_shexp.weight
+ 694: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_gate_shexp.weight
+ 695: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_up_shexp.weight
+ 696: 3072 | 3072, 1, 1, 1 | F32 | blk.37.ffn_gate_inp_shexp.weight
+ 697: 3072 | 3072, 1, 1, 1 | F32 | blk.37.post_attention_norm.weight
+ 698: 3072 | 3072, 1, 1, 1 | F32 | blk.38.attn_norm.weight
+ 699: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_a
+ 700: 49152 | 4, 12288, 1, 1 | F32 | blk.38.ssm_conv1d.weight
+ 701: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_dt.bias
+ 702: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_alpha.weight
+ 703: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_beta.weight
+ 704: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.38.attn_qkv.weight
+ 705: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.38.attn_gate.weight
+ 706: 128 | 128, 1, 1, 1 | F32 | blk.38.ssm_norm.weight
+ 707: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.38.ssm_out.weight
+ 708: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.38.ffn_down_exps.weight
+ 709: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.38.ffn_gate_exps.weight
+ 710: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.38.ffn_up_exps.weight
+ 711: 786432 | 3072, 256, 1, 1 | F32 | blk.38.ffn_gate_inp.weight
+ 712: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.38.ffn_down_shexp.weight
+ 713: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_gate_shexp.weight
+ 714: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_up_shexp.weight
+ 715: 3072 | 3072, 1, 1, 1 | F32 | blk.38.ffn_gate_inp_shexp.weight
+ 716: 3072 | 3072, 1, 1, 1 | F32 | blk.38.post_attention_norm.weight
+ 717: 3072 | 3072, 1, 1, 1 | F32 | blk.39.attn_norm.weight
+ 718: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.39.ffn_down_exps.weight
+ 719: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.39.ffn_gate_exps.weight
+ 720: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.39.ffn_up_exps.weight
+ 721: 786432 | 3072, 256, 1, 1 | F32 | blk.39.ffn_gate_inp.weight
+ 722: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.39.ffn_down_shexp.weight
+ 723: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_gate_shexp.weight
+ 724: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_up_shexp.weight
+ 725: 3072 | 3072, 1, 1, 1 | F32 | blk.39.ffn_gate_inp_shexp.weight
+ 726: 3072 | 3072, 1, 1, 1 | F32 | blk.39.post_attention_norm.weight
+ 727: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_k_norm.weight
+ 728: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_k.weight
+ 729: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.39.attn_output.weight
+ 730: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_q_norm.weight
+ 731: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.39.attn_q.weight
+ 732: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_v.weight
+ 733: 3072 | 3072, 1, 1, 1 | F32 | blk.40.attn_norm.weight
+ 734: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_a
+ 735: 49152 | 4, 12288, 1, 1 | F32 | blk.40.ssm_conv1d.weight
+ 736: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_dt.bias
+ 737: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_alpha.weight
+ 738: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_beta.weight
+ 739: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.40.attn_qkv.weight
+ 740: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.40.attn_gate.weight
+ 741: 128 | 128, 1, 1, 1 | F32 | blk.40.ssm_norm.weight
+ 742: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.40.ssm_out.weight
+ 743: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.40.ffn_down_exps.weight
+ 744: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.40.ffn_gate_exps.weight
+ 745: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.40.ffn_up_exps.weight
+ 746: 786432 | 3072, 256, 1, 1 | F32 | blk.40.ffn_gate_inp.weight
+ 747: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.40.ffn_down_shexp.weight
+ 748: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_gate_shexp.weight
+ 749: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_up_shexp.weight
+ 750: 3072 | 3072, 1, 1, 1 | F32 | blk.40.ffn_gate_inp_shexp.weight
+ 751: 3072 | 3072, 1, 1, 1 | F32 | blk.40.post_attention_norm.weight
+ 752: 3072 | 3072, 1, 1, 1 | F32 | blk.41.attn_norm.weight
+ 753: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_a
+ 754: 49152 | 4, 12288, 1, 1 | F32 | blk.41.ssm_conv1d.weight
+ 755: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_dt.bias
+ 756: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_alpha.weight
+ 757: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_beta.weight
+ 758: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.41.attn_qkv.weight
+ 759: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.41.attn_gate.weight
+ 760: 128 | 128, 1, 1, 1 | F32 | blk.41.ssm_norm.weight
+ 761: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.41.ssm_out.weight
+ 762: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.41.ffn_down_exps.weight
+ 763: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.41.ffn_gate_exps.weight
+ 764: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.41.ffn_up_exps.weight
+ 765: 786432 | 3072, 256, 1, 1 | F32 | blk.41.ffn_gate_inp.weight
+ 766: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.41.ffn_down_shexp.weight
+ 767: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_gate_shexp.weight
+ 768: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_up_shexp.weight
+ 769: 3072 | 3072, 1, 1, 1 | F32 | blk.41.ffn_gate_inp_shexp.weight
+ 770: 3072 | 3072, 1, 1, 1 | F32 | blk.41.post_attention_norm.weight
+ 771: 3072 | 3072, 1, 1, 1 | F32 | blk.42.attn_norm.weight
+ 772: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_a
+ 773: 49152 | 4, 12288, 1, 1 | F32 | blk.42.ssm_conv1d.weight
+ 774: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_dt.bias
+ 775: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_alpha.weight
+ 776: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_beta.weight
+ 777: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.42.attn_qkv.weight
+ 778: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.42.attn_gate.weight
+ 779: 128 | 128, 1, 1, 1 | F32 | blk.42.ssm_norm.weight
+ 780: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.42.ssm_out.weight
+ 781: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.42.ffn_down_exps.weight
+ 782: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.42.ffn_gate_exps.weight
+ 783: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.42.ffn_up_exps.weight
+ 784: 786432 | 3072, 256, 1, 1 | F32 | blk.42.ffn_gate_inp.weight
+ 785: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.42.ffn_down_shexp.weight
+ 786: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_gate_shexp.weight
+ 787: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_up_shexp.weight
+ 788: 3072 | 3072, 1, 1, 1 | F32 | blk.42.ffn_gate_inp_shexp.weight
+ 789: 3072 | 3072, 1, 1, 1 | F32 | blk.42.post_attention_norm.weight
+ 790: 3072 | 3072, 1, 1, 1 | F32 | blk.43.attn_norm.weight
+ 791: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.43.ffn_down_exps.weight
+ 792: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.43.ffn_gate_exps.weight
+ 793: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.43.ffn_up_exps.weight
+ 794: 786432 | 3072, 256, 1, 1 | F32 | blk.43.ffn_gate_inp.weight
+ 795: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.43.ffn_down_shexp.weight
+ 796: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_gate_shexp.weight
+ 797: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_up_shexp.weight
+ 798: 3072 | 3072, 1, 1, 1 | F32 | blk.43.ffn_gate_inp_shexp.weight
+ 799: 3072 | 3072, 1, 1, 1 | F32 | blk.43.post_attention_norm.weight
+ 800: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_k_norm.weight
+ 801: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_k.weight
+ 802: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.43.attn_output.weight
+ 803: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_q_norm.weight
+ 804: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.43.attn_q.weight
+ 805: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_v.weight
+ 806: 3072 | 3072, 1, 1, 1 | F32 | blk.44.attn_norm.weight
+ 807: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_a
+ 808: 49152 | 4, 12288, 1, 1 | F32 | blk.44.ssm_conv1d.weight
+ 809: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_dt.bias
+ 810: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_alpha.weight
+ 811: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_beta.weight
+ 812: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.44.attn_qkv.weight
+ 813: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.44.attn_gate.weight
+ 814: 128 | 128, 1, 1, 1 | F32 | blk.44.ssm_norm.weight
+ 815: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.44.ssm_out.weight
+ 816: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.44.ffn_down_exps.weight
+ 817: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.44.ffn_gate_exps.weight
+ 818: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.44.ffn_up_exps.weight
+ 819: 786432 | 3072, 256, 1, 1 | F32 | blk.44.ffn_gate_inp.weight
+ 820: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.44.ffn_down_shexp.weight
+ 821: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_gate_shexp.weight
+ 822: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_up_shexp.weight
+ 823: 3072 | 3072, 1, 1, 1 | F32 | blk.44.ffn_gate_inp_shexp.weight
+ 824: 3072 | 3072, 1, 1, 1 | F32 | blk.44.post_attention_norm.weight
+ 825: 3072 | 3072, 1, 1, 1 | F32 | blk.45.attn_norm.weight
+ 826: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_a
+ 827: 49152 | 4, 12288, 1, 1 | F32 | blk.45.ssm_conv1d.weight
+ 828: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_dt.bias
+ 829: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_alpha.weight
+ 830: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_beta.weight
+ 831: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.45.attn_qkv.weight
+ 832: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.45.attn_gate.weight
+ 833: 128 | 128, 1, 1, 1 | F32 | blk.45.ssm_norm.weight
+ 834: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.45.ssm_out.weight
+ 835: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.45.ffn_down_exps.weight
+ 836: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.45.ffn_gate_exps.weight
+ 837: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.45.ffn_up_exps.weight
+ 838: 786432 | 3072, 256, 1, 1 | F32 | blk.45.ffn_gate_inp.weight
+ 839: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.45.ffn_down_shexp.weight
+ 840: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_gate_shexp.weight
+ 841: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_up_shexp.weight
+ 842: 3072 | 3072, 1, 1, 1 | F32 | blk.45.ffn_gate_inp_shexp.weight
+ 843: 3072 | 3072, 1, 1, 1 | F32 | blk.45.post_attention_norm.weight
+ 844: 3072 | 3072, 1, 1, 1 | F32 | blk.46.attn_norm.weight
+ 845: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_a
+ 846: 49152 | 4, 12288, 1, 1 | F32 | blk.46.ssm_conv1d.weight
+ 847: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_dt.bias
+ 848: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_alpha.weight
+ 849: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_beta.weight
+ 850: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.46.attn_qkv.weight
+ 851: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.46.attn_gate.weight
+ 852: 128 | 128, 1, 1, 1 | F32 | blk.46.ssm_norm.weight
+ 853: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.46.ssm_out.weight
+ 854: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.46.ffn_down_exps.weight
+ 855: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.46.ffn_gate_exps.weight
+ 856: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.46.ffn_up_exps.weight
+ 857: 786432 | 3072, 256, 1, 1 | F32 | blk.46.ffn_gate_inp.weight
+ 858: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.46.ffn_down_shexp.weight
+ 859: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_gate_shexp.weight
+ 860: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_up_shexp.weight
+ 861: 3072 | 3072, 1, 1, 1 | F32 | blk.46.ffn_gate_inp_shexp.weight
+ 862: 3072 | 3072, 1, 1, 1 | F32 | blk.46.post_attention_norm.weight
+ 863: 3072 | 3072, 1, 1, 1 | F32 | blk.47.attn_norm.weight
+ 864: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.47.ffn_down_exps.weight
+ 865: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.47.ffn_gate_exps.weight
+ 866: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.47.ffn_up_exps.weight
+ 867: 786432 | 3072, 256, 1, 1 | F32 | blk.47.ffn_gate_inp.weight
+ 868: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.47.ffn_down_shexp.weight
+ 869: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_gate_shexp.weight
+ 870: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_up_shexp.weight
+ 871: 3072 | 3072, 1, 1, 1 | F32 | blk.47.ffn_gate_inp_shexp.weight
+ 872: 3072 | 3072, 1, 1, 1 | F32 | blk.47.post_attention_norm.weight
+ 873: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_k_norm.weight
+ 874: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_k.weight
+ 875: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.47.attn_output.weight
+ 876: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_q_norm.weight
+ 877: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.47.attn_q.weight
+ 878: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_v.weight
+ 879: 3072 | 3072, 1, 1, 1 | F32 | output_norm.weight
+
+
+
+**MQ68 (Mixed Quant Q6_K - Q8_0)**
+
+gguf-dump
+
+ --leave-output-tensor
+ --tensor-type ".*ffn_down_exps\.weight=q6_k"
+ --tensor-type ".*ffn_gate_exps\.weight=q6_k"
+ --tensor-type ".*ffn_up_exps\.weight=q6_k"
+ Q8_0
+
+ 1: UINT32 | 1 | GGUF.version = 3
+ 2: UINT64 | 1 | GGUF.tensor_count = 879
+ 3: UINT64 | 1 | GGUF.kv_count = 43
+ 4: STRING | 1 | general.architecture = 'qwen35moe'
+ 5: STRING | 1 | general.type = 'model'
+ 6: INT32 | 1 | general.sampling.top_k = 20
+ 7: FLOAT32 | 1 | general.sampling.top_p = 0.949999988079071
+ 8: FLOAT32 | 1 | general.sampling.temp = 0.6000000238418579
+ 9: STRING | 1 | general.name = 'Qwen3.5 122B A10B Wangzhang Abliterix No MTP'
+ 10: STRING | 1 | general.size_label = '256x6.6B'
+ 11: UINT32 | 1 | general.base_model.count = 1
+ 12: STRING | 1 | general.base_model.0.name = 'Qwen3.5 122B A10B'
+ 13: STRING | 1 | general.base_model.0.organization = 'Qwen'
+ 14: STRING | 1 | general.base_model.0.repo_url = 'https://huggingface.co/Qwen/Qwen3.5-122B-A10B'
+ 15: [STRING] | 5 | general.tags = ['abliterix', 'uncensored', 'decensored', 'abliterated', 'text-generation']
+ 16: UINT32 | 1 | qwen35moe.block_count = 48
+ 17: UINT32 | 1 | qwen35moe.context_length = 262144
+ 18: UINT32 | 1 | qwen35moe.embedding_length = 3072
+ 19: UINT32 | 1 | qwen35moe.attention.head_count = 32
+ 20: UINT32 | 1 | qwen35moe.attention.head_count_kv = 2
+ 21: [INT32] | 4 | qwen35moe.rope.dimension_sections = [11, 11, 10, 0]
+ 22: FLOAT32 | 1 | qwen35moe.rope.freq_base = 10000000.0
+ 23: FLOAT32 | 1 | qwen35moe.attention.layer_norm_rms_epsilon = 9.999999974752427e-07
+ 24: UINT32 | 1 | qwen35moe.expert_count = 256
+ 25: UINT32 | 1 | qwen35moe.expert_used_count = 8
+ 26: UINT32 | 1 | qwen35moe.attention.key_length = 256
+ 27: UINT32 | 1 | qwen35moe.attention.value_length = 256
+ 28: UINT32 | 1 | qwen35moe.expert_feed_forward_length = 1024
+ 29: UINT32 | 1 | qwen35moe.expert_shared_feed_forward_length = 1024
+ 30: UINT32 | 1 | qwen35moe.ssm.conv_kernel = 4
+ 31: UINT32 | 1 | qwen35moe.ssm.state_size = 128
+ 32: UINT32 | 1 | qwen35moe.ssm.group_count = 16
+ 33: UINT32 | 1 | qwen35moe.ssm.time_step_rank = 64
+ 34: UINT32 | 1 | qwen35moe.ssm.inner_size = 8192
+ 35: UINT32 | 1 | qwen35moe.full_attention_interval = 4
+ 36: UINT32 | 1 | qwen35moe.rope.dimension_count = 64
+ 37: STRING | 1 | tokenizer.ggml.model = 'gpt2'
+ 38: STRING | 1 | tokenizer.ggml.pre = 'qwen35'
+ 39: [STRING] | 248320 | tokenizer.ggml.tokens = ['!', '"', '#', '$', '%', '&', ...]
+ 40: [INT32] | 248320 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...]
+ 41: [STRING] | 247587 | tokenizer.ggml.merges = ['Ġ Ġ', 'ĠĠ ĠĠ', 'i n', 'Ġ t', 'ĠĠĠĠ ĠĠĠĠ', 'e r', ...]
+ 42: UINT32 | 1 | tokenizer.ggml.eos_token_id = 248046
+ 43: UINT32 | 1 | tokenizer.ggml.padding_token_id = 248044
+ 44: STRING | 1 | tokenizer.chat_template = '{%- set image_count = namespace(value=0) %}\n{%- set video...'
+ 45: UINT32 | 1 | general.quantization_version = 2
+ 46: UINT32 | 1 | general.file_type = 7
+
+ 1: 762839040 | 3072, 248320, 1, 1 | BF16 | output.weight
+ 2: 3072 | 3072, 1, 1, 1 | F32 | output_norm.weight
+ 3: 762839040 | 3072, 248320, 1, 1 | Q8_0 | token_embd.weight
+ 4: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.0.attn_gate.weight
+ 5: 3072 | 3072, 1, 1, 1 | F32 | blk.0.attn_norm.weight
+ 6: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.0.attn_qkv.weight
+ 7: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.0.ffn_down_exps.weight
+ 8: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.0.ffn_down_shexp.weight
+ 9: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.0.ffn_gate_exps.weight
+ 10: 786432 | 3072, 256, 1, 1 | F32 | blk.0.ffn_gate_inp.weight
+ 11: 3072 | 3072, 1, 1, 1 | F32 | blk.0.ffn_gate_inp_shexp.weight
+ 12: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.0.ffn_gate_shexp.weight
+ 13: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.0.ffn_up_exps.weight
+ 14: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.0.ffn_up_shexp.weight
+ 15: 3072 | 3072, 1, 1, 1 | F32 | blk.0.post_attention_norm.weight
+ 16: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_a
+ 17: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.0.ssm_alpha.weight
+ 18: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.0.ssm_beta.weight
+ 19: 49152 | 4, 12288, 1, 1 | F32 | blk.0.ssm_conv1d.weight
+ 20: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_dt.bias
+ 21: 128 | 128, 1, 1, 1 | F32 | blk.0.ssm_norm.weight
+ 22: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.0.ssm_out.weight
+ 23: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.1.attn_gate.weight
+ 24: 3072 | 3072, 1, 1, 1 | F32 | blk.1.attn_norm.weight
+ 25: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.1.attn_qkv.weight
+ 26: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.1.ffn_down_exps.weight
+ 27: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.1.ffn_down_shexp.weight
+ 28: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.1.ffn_gate_exps.weight
+ 29: 786432 | 3072, 256, 1, 1 | F32 | blk.1.ffn_gate_inp.weight
+ 30: 3072 | 3072, 1, 1, 1 | F32 | blk.1.ffn_gate_inp_shexp.weight
+ 31: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.1.ffn_gate_shexp.weight
+ 32: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.1.ffn_up_exps.weight
+ 33: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.1.ffn_up_shexp.weight
+ 34: 3072 | 3072, 1, 1, 1 | F32 | blk.1.post_attention_norm.weight
+ 35: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_a
+ 36: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.1.ssm_alpha.weight
+ 37: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.1.ssm_beta.weight
+ 38: 49152 | 4, 12288, 1, 1 | F32 | blk.1.ssm_conv1d.weight
+ 39: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_dt.bias
+ 40: 128 | 128, 1, 1, 1 | F32 | blk.1.ssm_norm.weight
+ 41: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.1.ssm_out.weight
+ 42: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.2.attn_gate.weight
+ 43: 3072 | 3072, 1, 1, 1 | F32 | blk.2.attn_norm.weight
+ 44: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.2.attn_qkv.weight
+ 45: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.2.ffn_down_exps.weight
+ 46: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.2.ffn_down_shexp.weight
+ 47: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.2.ffn_gate_exps.weight
+ 48: 786432 | 3072, 256, 1, 1 | F32 | blk.2.ffn_gate_inp.weight
+ 49: 3072 | 3072, 1, 1, 1 | F32 | blk.2.ffn_gate_inp_shexp.weight
+ 50: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.2.ffn_gate_shexp.weight
+ 51: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.2.ffn_up_exps.weight
+ 52: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.2.ffn_up_shexp.weight
+ 53: 3072 | 3072, 1, 1, 1 | F32 | blk.2.post_attention_norm.weight
+ 54: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_a
+ 55: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.2.ssm_alpha.weight
+ 56: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.2.ssm_beta.weight
+ 57: 49152 | 4, 12288, 1, 1 | F32 | blk.2.ssm_conv1d.weight
+ 58: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_dt.bias
+ 59: 128 | 128, 1, 1, 1 | F32 | blk.2.ssm_norm.weight
+ 60: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.2.ssm_out.weight
+ 61: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.3.attn_k.weight
+ 62: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_k_norm.weight
+ 63: 3072 | 3072, 1, 1, 1 | F32 | blk.3.attn_norm.weight
+ 64: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.3.attn_output.weight
+ 65: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.3.attn_q.weight
+ 66: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_q_norm.weight
+ 67: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.3.attn_v.weight
+ 68: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.3.ffn_down_exps.weight
+ 69: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.3.ffn_down_shexp.weight
+ 70: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.3.ffn_gate_exps.weight
+ 71: 786432 | 3072, 256, 1, 1 | F32 | blk.3.ffn_gate_inp.weight
+ 72: 3072 | 3072, 1, 1, 1 | F32 | blk.3.ffn_gate_inp_shexp.weight
+ 73: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.3.ffn_gate_shexp.weight
+ 74: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.3.ffn_up_exps.weight
+ 75: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.3.ffn_up_shexp.weight
+ 76: 3072 | 3072, 1, 1, 1 | F32 | blk.3.post_attention_norm.weight
+ 77: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.4.attn_gate.weight
+ 78: 3072 | 3072, 1, 1, 1 | F32 | blk.4.attn_norm.weight
+ 79: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.4.attn_qkv.weight
+ 80: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.4.ffn_down_exps.weight
+ 81: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.4.ffn_down_shexp.weight
+ 82: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.4.ffn_gate_exps.weight
+ 83: 786432 | 3072, 256, 1, 1 | F32 | blk.4.ffn_gate_inp.weight
+ 84: 3072 | 3072, 1, 1, 1 | F32 | blk.4.ffn_gate_inp_shexp.weight
+ 85: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.4.ffn_gate_shexp.weight
+ 86: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.4.ffn_up_exps.weight
+ 87: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.4.ffn_up_shexp.weight
+ 88: 3072 | 3072, 1, 1, 1 | F32 | blk.4.post_attention_norm.weight
+ 89: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_a
+ 90: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.4.ssm_alpha.weight
+ 91: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.4.ssm_beta.weight
+ 92: 49152 | 4, 12288, 1, 1 | F32 | blk.4.ssm_conv1d.weight
+ 93: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_dt.bias
+ 94: 128 | 128, 1, 1, 1 | F32 | blk.4.ssm_norm.weight
+ 95: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.4.ssm_out.weight
+ 96: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.5.attn_gate.weight
+ 97: 3072 | 3072, 1, 1, 1 | F32 | blk.5.attn_norm.weight
+ 98: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.5.attn_qkv.weight
+ 99: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.5.ffn_down_exps.weight
+ 100: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.5.ffn_down_shexp.weight
+ 101: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.5.ffn_gate_exps.weight
+ 102: 786432 | 3072, 256, 1, 1 | F32 | blk.5.ffn_gate_inp.weight
+ 103: 3072 | 3072, 1, 1, 1 | F32 | blk.5.ffn_gate_inp_shexp.weight
+ 104: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.5.ffn_gate_shexp.weight
+ 105: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.5.ffn_up_exps.weight
+ 106: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.5.ffn_up_shexp.weight
+ 107: 3072 | 3072, 1, 1, 1 | F32 | blk.5.post_attention_norm.weight
+ 108: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_a
+ 109: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.5.ssm_alpha.weight
+ 110: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.5.ssm_beta.weight
+ 111: 49152 | 4, 12288, 1, 1 | F32 | blk.5.ssm_conv1d.weight
+ 112: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_dt.bias
+ 113: 128 | 128, 1, 1, 1 | F32 | blk.5.ssm_norm.weight
+ 114: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.5.ssm_out.weight
+ 115: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.6.attn_gate.weight
+ 116: 3072 | 3072, 1, 1, 1 | F32 | blk.6.attn_norm.weight
+ 117: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.6.attn_qkv.weight
+ 118: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.6.ffn_down_exps.weight
+ 119: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.6.ffn_down_shexp.weight
+ 120: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.6.ffn_gate_exps.weight
+ 121: 786432 | 3072, 256, 1, 1 | F32 | blk.6.ffn_gate_inp.weight
+ 122: 3072 | 3072, 1, 1, 1 | F32 | blk.6.ffn_gate_inp_shexp.weight
+ 123: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.6.ffn_gate_shexp.weight
+ 124: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.6.ffn_up_exps.weight
+ 125: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.6.ffn_up_shexp.weight
+ 126: 3072 | 3072, 1, 1, 1 | F32 | blk.6.post_attention_norm.weight
+ 127: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_a
+ 128: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.6.ssm_alpha.weight
+ 129: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.6.ssm_beta.weight
+ 130: 49152 | 4, 12288, 1, 1 | F32 | blk.6.ssm_conv1d.weight
+ 131: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_dt.bias
+ 132: 128 | 128, 1, 1, 1 | F32 | blk.6.ssm_norm.weight
+ 133: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.6.ssm_out.weight
+ 134: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.7.attn_k.weight
+ 135: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_k_norm.weight
+ 136: 3072 | 3072, 1, 1, 1 | F32 | blk.7.attn_norm.weight
+ 137: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.7.attn_output.weight
+ 138: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.7.attn_q.weight
+ 139: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_q_norm.weight
+ 140: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.7.attn_v.weight
+ 141: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.7.ffn_down_exps.weight
+ 142: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.7.ffn_down_shexp.weight
+ 143: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.7.ffn_gate_exps.weight
+ 144: 786432 | 3072, 256, 1, 1 | F32 | blk.7.ffn_gate_inp.weight
+ 145: 3072 | 3072, 1, 1, 1 | F32 | blk.7.ffn_gate_inp_shexp.weight
+ 146: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.7.ffn_gate_shexp.weight
+ 147: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.7.ffn_up_exps.weight
+ 148: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.7.ffn_up_shexp.weight
+ 149: 3072 | 3072, 1, 1, 1 | F32 | blk.7.post_attention_norm.weight
+ 150: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.8.attn_gate.weight
+ 151: 3072 | 3072, 1, 1, 1 | F32 | blk.8.attn_norm.weight
+ 152: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.8.attn_qkv.weight
+ 153: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.8.ffn_down_exps.weight
+ 154: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.8.ffn_down_shexp.weight
+ 155: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.8.ffn_gate_exps.weight
+ 156: 786432 | 3072, 256, 1, 1 | F32 | blk.8.ffn_gate_inp.weight
+ 157: 3072 | 3072, 1, 1, 1 | F32 | blk.8.ffn_gate_inp_shexp.weight
+ 158: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.8.ffn_gate_shexp.weight
+ 159: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.8.ffn_up_exps.weight
+ 160: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.8.ffn_up_shexp.weight
+ 161: 3072 | 3072, 1, 1, 1 | F32 | blk.8.post_attention_norm.weight
+ 162: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_a
+ 163: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.8.ssm_alpha.weight
+ 164: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.8.ssm_beta.weight
+ 165: 49152 | 4, 12288, 1, 1 | F32 | blk.8.ssm_conv1d.weight
+ 166: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_dt.bias
+ 167: 128 | 128, 1, 1, 1 | F32 | blk.8.ssm_norm.weight
+ 168: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.8.ssm_out.weight
+ 169: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.9.attn_gate.weight
+ 170: 3072 | 3072, 1, 1, 1 | F32 | blk.9.attn_norm.weight
+ 171: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.9.attn_qkv.weight
+ 172: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.9.ffn_down_exps.weight
+ 173: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.9.ffn_down_shexp.weight
+ 174: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.9.ffn_gate_exps.weight
+ 175: 786432 | 3072, 256, 1, 1 | F32 | blk.9.ffn_gate_inp.weight
+ 176: 3072 | 3072, 1, 1, 1 | F32 | blk.9.ffn_gate_inp_shexp.weight
+ 177: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.9.ffn_gate_shexp.weight
+ 178: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.9.ffn_up_exps.weight
+ 179: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.9.ffn_up_shexp.weight
+ 180: 3072 | 3072, 1, 1, 1 | F32 | blk.9.post_attention_norm.weight
+ 181: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_a
+ 182: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.9.ssm_alpha.weight
+ 183: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.9.ssm_beta.weight
+ 184: 49152 | 4, 12288, 1, 1 | F32 | blk.9.ssm_conv1d.weight
+ 185: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_dt.bias
+ 186: 128 | 128, 1, 1, 1 | F32 | blk.9.ssm_norm.weight
+ 187: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.9.ssm_out.weight
+ 188: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.10.attn_gate.weight
+ 189: 3072 | 3072, 1, 1, 1 | F32 | blk.10.attn_norm.weight
+ 190: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.10.attn_qkv.weight
+ 191: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.10.ffn_down_exps.weight
+ 192: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.10.ffn_down_shexp.weight
+ 193: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.10.ffn_gate_exps.weight
+ 194: 786432 | 3072, 256, 1, 1 | F32 | blk.10.ffn_gate_inp.weight
+ 195: 3072 | 3072, 1, 1, 1 | F32 | blk.10.ffn_gate_inp_shexp.weight
+ 196: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.10.ffn_gate_shexp.weight
+ 197: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.10.ffn_up_exps.weight
+ 198: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.10.ffn_up_shexp.weight
+ 199: 3072 | 3072, 1, 1, 1 | F32 | blk.10.post_attention_norm.weight
+ 200: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_a
+ 201: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.10.ssm_alpha.weight
+ 202: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.10.ssm_beta.weight
+ 203: 49152 | 4, 12288, 1, 1 | F32 | blk.10.ssm_conv1d.weight
+ 204: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_dt.bias
+ 205: 128 | 128, 1, 1, 1 | F32 | blk.10.ssm_norm.weight
+ 206: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.10.ssm_out.weight
+ 207: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.11.attn_k.weight
+ 208: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_k_norm.weight
+ 209: 3072 | 3072, 1, 1, 1 | F32 | blk.11.attn_norm.weight
+ 210: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.11.attn_output.weight
+ 211: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.11.attn_q.weight
+ 212: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_q_norm.weight
+ 213: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.11.attn_v.weight
+ 214: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.11.ffn_down_exps.weight
+ 215: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.11.ffn_down_shexp.weight
+ 216: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.11.ffn_gate_exps.weight
+ 217: 786432 | 3072, 256, 1, 1 | F32 | blk.11.ffn_gate_inp.weight
+ 218: 3072 | 3072, 1, 1, 1 | F32 | blk.11.ffn_gate_inp_shexp.weight
+ 219: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.11.ffn_gate_shexp.weight
+ 220: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.11.ffn_up_exps.weight
+ 221: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.11.ffn_up_shexp.weight
+ 222: 3072 | 3072, 1, 1, 1 | F32 | blk.11.post_attention_norm.weight
+ 223: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.12.attn_gate.weight
+ 224: 3072 | 3072, 1, 1, 1 | F32 | blk.12.attn_norm.weight
+ 225: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.12.attn_qkv.weight
+ 226: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.12.ffn_down_exps.weight
+ 227: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.12.ffn_down_shexp.weight
+ 228: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.12.ffn_gate_exps.weight
+ 229: 786432 | 3072, 256, 1, 1 | F32 | blk.12.ffn_gate_inp.weight
+ 230: 3072 | 3072, 1, 1, 1 | F32 | blk.12.ffn_gate_inp_shexp.weight
+ 231: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.12.ffn_gate_shexp.weight
+ 232: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.12.ffn_up_exps.weight
+ 233: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.12.ffn_up_shexp.weight
+ 234: 3072 | 3072, 1, 1, 1 | F32 | blk.12.post_attention_norm.weight
+ 235: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_a
+ 236: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.12.ssm_alpha.weight
+ 237: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.12.ssm_beta.weight
+ 238: 49152 | 4, 12288, 1, 1 | F32 | blk.12.ssm_conv1d.weight
+ 239: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_dt.bias
+ 240: 128 | 128, 1, 1, 1 | F32 | blk.12.ssm_norm.weight
+ 241: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.12.ssm_out.weight
+ 242: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.13.attn_gate.weight
+ 243: 3072 | 3072, 1, 1, 1 | F32 | blk.13.attn_norm.weight
+ 244: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.13.attn_qkv.weight
+ 245: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.13.ffn_down_exps.weight
+ 246: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.13.ffn_down_shexp.weight
+ 247: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.13.ffn_gate_exps.weight
+ 248: 786432 | 3072, 256, 1, 1 | F32 | blk.13.ffn_gate_inp.weight
+ 249: 3072 | 3072, 1, 1, 1 | F32 | blk.13.ffn_gate_inp_shexp.weight
+ 250: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.13.ffn_gate_shexp.weight
+ 251: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.13.ffn_up_exps.weight
+ 252: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.13.ffn_up_shexp.weight
+ 253: 3072 | 3072, 1, 1, 1 | F32 | blk.13.post_attention_norm.weight
+ 254: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_a
+ 255: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.13.ssm_alpha.weight
+ 256: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.13.ssm_beta.weight
+ 257: 49152 | 4, 12288, 1, 1 | F32 | blk.13.ssm_conv1d.weight
+ 258: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_dt.bias
+ 259: 128 | 128, 1, 1, 1 | F32 | blk.13.ssm_norm.weight
+ 260: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.13.ssm_out.weight
+ 261: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.14.attn_gate.weight
+ 262: 3072 | 3072, 1, 1, 1 | F32 | blk.14.attn_norm.weight
+ 263: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.14.attn_qkv.weight
+ 264: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.14.ffn_down_exps.weight
+ 265: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.14.ffn_down_shexp.weight
+ 266: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.14.ffn_gate_exps.weight
+ 267: 786432 | 3072, 256, 1, 1 | F32 | blk.14.ffn_gate_inp.weight
+ 268: 3072 | 3072, 1, 1, 1 | F32 | blk.14.ffn_gate_inp_shexp.weight
+ 269: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.14.ffn_gate_shexp.weight
+ 270: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.14.ffn_up_exps.weight
+ 271: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.14.ffn_up_shexp.weight
+ 272: 3072 | 3072, 1, 1, 1 | F32 | blk.14.post_attention_norm.weight
+ 273: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_a
+ 274: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.14.ssm_alpha.weight
+ 275: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.14.ssm_beta.weight
+ 276: 49152 | 4, 12288, 1, 1 | F32 | blk.14.ssm_conv1d.weight
+ 277: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_dt.bias
+ 278: 128 | 128, 1, 1, 1 | F32 | blk.14.ssm_norm.weight
+ 279: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.14.ssm_out.weight
+ 280: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.15.attn_k.weight
+ 281: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_k_norm.weight
+ 282: 3072 | 3072, 1, 1, 1 | F32 | blk.15.attn_norm.weight
+ 283: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.15.attn_output.weight
+ 284: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.15.attn_q.weight
+ 285: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_q_norm.weight
+ 286: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.15.attn_v.weight
+ 287: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.15.ffn_down_exps.weight
+ 288: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.15.ffn_down_shexp.weight
+ 289: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.15.ffn_gate_exps.weight
+ 290: 786432 | 3072, 256, 1, 1 | F32 | blk.15.ffn_gate_inp.weight
+ 291: 3072 | 3072, 1, 1, 1 | F32 | blk.15.ffn_gate_inp_shexp.weight
+ 292: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.15.ffn_gate_shexp.weight
+ 293: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.15.ffn_up_exps.weight
+ 294: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.15.ffn_up_shexp.weight
+ 295: 3072 | 3072, 1, 1, 1 | F32 | blk.15.post_attention_norm.weight
+ 296: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.16.attn_gate.weight
+ 297: 3072 | 3072, 1, 1, 1 | F32 | blk.16.attn_norm.weight
+ 298: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.16.attn_qkv.weight
+ 299: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.16.ffn_down_exps.weight
+ 300: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.16.ffn_down_shexp.weight
+ 301: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.16.ffn_gate_exps.weight
+ 302: 786432 | 3072, 256, 1, 1 | F32 | blk.16.ffn_gate_inp.weight
+ 303: 3072 | 3072, 1, 1, 1 | F32 | blk.16.ffn_gate_inp_shexp.weight
+ 304: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.16.ffn_gate_shexp.weight
+ 305: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.16.ffn_up_exps.weight
+ 306: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.16.ffn_up_shexp.weight
+ 307: 3072 | 3072, 1, 1, 1 | F32 | blk.16.post_attention_norm.weight
+ 308: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_a
+ 309: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.16.ssm_alpha.weight
+ 310: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.16.ssm_beta.weight
+ 311: 49152 | 4, 12288, 1, 1 | F32 | blk.16.ssm_conv1d.weight
+ 312: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_dt.bias
+ 313: 128 | 128, 1, 1, 1 | F32 | blk.16.ssm_norm.weight
+ 314: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.16.ssm_out.weight
+ 315: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.17.attn_gate.weight
+ 316: 3072 | 3072, 1, 1, 1 | F32 | blk.17.attn_norm.weight
+ 317: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.17.attn_qkv.weight
+ 318: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.17.ffn_down_exps.weight
+ 319: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.17.ffn_down_shexp.weight
+ 320: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.17.ffn_gate_exps.weight
+ 321: 786432 | 3072, 256, 1, 1 | F32 | blk.17.ffn_gate_inp.weight
+ 322: 3072 | 3072, 1, 1, 1 | F32 | blk.17.ffn_gate_inp_shexp.weight
+ 323: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.17.ffn_gate_shexp.weight
+ 324: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.17.ffn_up_exps.weight
+ 325: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.17.ffn_up_shexp.weight
+ 326: 3072 | 3072, 1, 1, 1 | F32 | blk.17.post_attention_norm.weight
+ 327: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_a
+ 328: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.17.ssm_alpha.weight
+ 329: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.17.ssm_beta.weight
+ 330: 49152 | 4, 12288, 1, 1 | F32 | blk.17.ssm_conv1d.weight
+ 331: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_dt.bias
+ 332: 128 | 128, 1, 1, 1 | F32 | blk.17.ssm_norm.weight
+ 333: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.17.ssm_out.weight
+ 334: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.18.attn_gate.weight
+ 335: 3072 | 3072, 1, 1, 1 | F32 | blk.18.attn_norm.weight
+ 336: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.18.attn_qkv.weight
+ 337: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.18.ffn_down_exps.weight
+ 338: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.18.ffn_down_shexp.weight
+ 339: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.18.ffn_gate_exps.weight
+ 340: 786432 | 3072, 256, 1, 1 | F32 | blk.18.ffn_gate_inp.weight
+ 341: 3072 | 3072, 1, 1, 1 | F32 | blk.18.ffn_gate_inp_shexp.weight
+ 342: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.18.ffn_gate_shexp.weight
+ 343: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.18.ffn_up_exps.weight
+ 344: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.18.ffn_up_shexp.weight
+ 345: 3072 | 3072, 1, 1, 1 | F32 | blk.18.post_attention_norm.weight
+ 346: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_a
+ 347: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.18.ssm_alpha.weight
+ 348: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.18.ssm_beta.weight
+ 349: 49152 | 4, 12288, 1, 1 | F32 | blk.18.ssm_conv1d.weight
+ 350: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_dt.bias
+ 351: 128 | 128, 1, 1, 1 | F32 | blk.18.ssm_norm.weight
+ 352: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.18.ssm_out.weight
+ 353: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.19.attn_k.weight
+ 354: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_k_norm.weight
+ 355: 3072 | 3072, 1, 1, 1 | F32 | blk.19.attn_norm.weight
+ 356: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.19.attn_output.weight
+ 357: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.19.attn_q.weight
+ 358: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_q_norm.weight
+ 359: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.19.attn_v.weight
+ 360: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.19.ffn_down_exps.weight
+ 361: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.19.ffn_down_shexp.weight
+ 362: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.19.ffn_gate_exps.weight
+ 363: 786432 | 3072, 256, 1, 1 | F32 | blk.19.ffn_gate_inp.weight
+ 364: 3072 | 3072, 1, 1, 1 | F32 | blk.19.ffn_gate_inp_shexp.weight
+ 365: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.19.ffn_gate_shexp.weight
+ 366: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.19.ffn_up_exps.weight
+ 367: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.19.ffn_up_shexp.weight
+ 368: 3072 | 3072, 1, 1, 1 | F32 | blk.19.post_attention_norm.weight
+ 369: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.20.attn_gate.weight
+ 370: 3072 | 3072, 1, 1, 1 | F32 | blk.20.attn_norm.weight
+ 371: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.20.attn_qkv.weight
+ 372: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.20.ffn_down_exps.weight
+ 373: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.20.ffn_down_shexp.weight
+ 374: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.20.ffn_gate_exps.weight
+ 375: 786432 | 3072, 256, 1, 1 | F32 | blk.20.ffn_gate_inp.weight
+ 376: 3072 | 3072, 1, 1, 1 | F32 | blk.20.ffn_gate_inp_shexp.weight
+ 377: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.20.ffn_gate_shexp.weight
+ 378: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.20.ffn_up_exps.weight
+ 379: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.20.ffn_up_shexp.weight
+ 380: 3072 | 3072, 1, 1, 1 | F32 | blk.20.post_attention_norm.weight
+ 381: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_a
+ 382: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.20.ssm_alpha.weight
+ 383: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.20.ssm_beta.weight
+ 384: 49152 | 4, 12288, 1, 1 | F32 | blk.20.ssm_conv1d.weight
+ 385: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_dt.bias
+ 386: 128 | 128, 1, 1, 1 | F32 | blk.20.ssm_norm.weight
+ 387: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.20.ssm_out.weight
+ 388: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.21.attn_gate.weight
+ 389: 3072 | 3072, 1, 1, 1 | F32 | blk.21.attn_norm.weight
+ 390: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.21.attn_qkv.weight
+ 391: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.21.ffn_down_exps.weight
+ 392: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.21.ffn_down_shexp.weight
+ 393: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.21.ffn_gate_exps.weight
+ 394: 786432 | 3072, 256, 1, 1 | F32 | blk.21.ffn_gate_inp.weight
+ 395: 3072 | 3072, 1, 1, 1 | F32 | blk.21.ffn_gate_inp_shexp.weight
+ 396: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.21.ffn_gate_shexp.weight
+ 397: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.21.ffn_up_exps.weight
+ 398: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.21.ffn_up_shexp.weight
+ 399: 3072 | 3072, 1, 1, 1 | F32 | blk.21.post_attention_norm.weight
+ 400: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_a
+ 401: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.21.ssm_alpha.weight
+ 402: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.21.ssm_beta.weight
+ 403: 49152 | 4, 12288, 1, 1 | F32 | blk.21.ssm_conv1d.weight
+ 404: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_dt.bias
+ 405: 128 | 128, 1, 1, 1 | F32 | blk.21.ssm_norm.weight
+ 406: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.21.ssm_out.weight
+ 407: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.22.attn_gate.weight
+ 408: 3072 | 3072, 1, 1, 1 | F32 | blk.22.attn_norm.weight
+ 409: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.22.attn_qkv.weight
+ 410: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.22.ffn_down_exps.weight
+ 411: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.22.ffn_down_shexp.weight
+ 412: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.22.ffn_gate_exps.weight
+ 413: 786432 | 3072, 256, 1, 1 | F32 | blk.22.ffn_gate_inp.weight
+ 414: 3072 | 3072, 1, 1, 1 | F32 | blk.22.ffn_gate_inp_shexp.weight
+ 415: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.22.ffn_gate_shexp.weight
+ 416: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.22.ffn_up_exps.weight
+ 417: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.22.ffn_up_shexp.weight
+ 418: 3072 | 3072, 1, 1, 1 | F32 | blk.22.post_attention_norm.weight
+ 419: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_a
+ 420: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.22.ssm_alpha.weight
+ 421: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.22.ssm_beta.weight
+ 422: 49152 | 4, 12288, 1, 1 | F32 | blk.22.ssm_conv1d.weight
+ 423: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_dt.bias
+ 424: 128 | 128, 1, 1, 1 | F32 | blk.22.ssm_norm.weight
+ 425: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.22.ssm_out.weight
+ 426: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.23.attn_k.weight
+ 427: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_k_norm.weight
+ 428: 3072 | 3072, 1, 1, 1 | F32 | blk.23.attn_norm.weight
+ 429: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.23.attn_output.weight
+ 430: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.23.attn_q.weight
+ 431: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_q_norm.weight
+ 432: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.23.attn_v.weight
+ 433: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.23.ffn_down_exps.weight
+ 434: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.23.ffn_down_shexp.weight
+ 435: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.23.ffn_gate_exps.weight
+ 436: 786432 | 3072, 256, 1, 1 | F32 | blk.23.ffn_gate_inp.weight
+ 437: 3072 | 3072, 1, 1, 1 | F32 | blk.23.ffn_gate_inp_shexp.weight
+ 438: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.23.ffn_gate_shexp.weight
+ 439: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.23.ffn_up_exps.weight
+ 440: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.23.ffn_up_shexp.weight
+ 441: 3072 | 3072, 1, 1, 1 | F32 | blk.23.post_attention_norm.weight
+ 442: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.24.attn_gate.weight
+ 443: 3072 | 3072, 1, 1, 1 | F32 | blk.24.attn_norm.weight
+ 444: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.24.attn_qkv.weight
+ 445: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.24.ffn_down_exps.weight
+ 446: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.24.ffn_down_shexp.weight
+ 447: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.24.ffn_gate_exps.weight
+ 448: 786432 | 3072, 256, 1, 1 | F32 | blk.24.ffn_gate_inp.weight
+ 449: 3072 | 3072, 1, 1, 1 | F32 | blk.24.ffn_gate_inp_shexp.weight
+ 450: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.24.ffn_gate_shexp.weight
+ 451: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.24.ffn_up_exps.weight
+ 452: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.24.ffn_up_shexp.weight
+ 453: 3072 | 3072, 1, 1, 1 | F32 | blk.24.post_attention_norm.weight
+ 454: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_a
+ 455: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.24.ssm_alpha.weight
+ 456: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.24.ssm_beta.weight
+ 457: 49152 | 4, 12288, 1, 1 | F32 | blk.24.ssm_conv1d.weight
+ 458: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_dt.bias
+ 459: 128 | 128, 1, 1, 1 | F32 | blk.24.ssm_norm.weight
+ 460: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.24.ssm_out.weight
+ 461: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.25.attn_gate.weight
+ 462: 3072 | 3072, 1, 1, 1 | F32 | blk.25.attn_norm.weight
+ 463: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.25.attn_qkv.weight
+ 464: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.25.ffn_down_exps.weight
+ 465: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.25.ffn_down_shexp.weight
+ 466: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.25.ffn_gate_exps.weight
+ 467: 786432 | 3072, 256, 1, 1 | F32 | blk.25.ffn_gate_inp.weight
+ 468: 3072 | 3072, 1, 1, 1 | F32 | blk.25.ffn_gate_inp_shexp.weight
+ 469: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.25.ffn_gate_shexp.weight
+ 470: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.25.ffn_up_exps.weight
+ 471: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.25.ffn_up_shexp.weight
+ 472: 3072 | 3072, 1, 1, 1 | F32 | blk.25.post_attention_norm.weight
+ 473: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_a
+ 474: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.25.ssm_alpha.weight
+ 475: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.25.ssm_beta.weight
+ 476: 49152 | 4, 12288, 1, 1 | F32 | blk.25.ssm_conv1d.weight
+ 477: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_dt.bias
+ 478: 128 | 128, 1, 1, 1 | F32 | blk.25.ssm_norm.weight
+ 479: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.25.ssm_out.weight
+ 480: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.26.attn_gate.weight
+ 481: 3072 | 3072, 1, 1, 1 | F32 | blk.26.attn_norm.weight
+ 482: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.26.attn_qkv.weight
+ 483: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.26.ffn_down_exps.weight
+ 484: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.26.ffn_down_shexp.weight
+ 485: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.26.ffn_gate_exps.weight
+ 486: 786432 | 3072, 256, 1, 1 | F32 | blk.26.ffn_gate_inp.weight
+ 487: 3072 | 3072, 1, 1, 1 | F32 | blk.26.ffn_gate_inp_shexp.weight
+ 488: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.26.ffn_gate_shexp.weight
+ 489: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.26.ffn_up_exps.weight
+ 490: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.26.ffn_up_shexp.weight
+ 491: 3072 | 3072, 1, 1, 1 | F32 | blk.26.post_attention_norm.weight
+ 492: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_a
+ 493: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.26.ssm_alpha.weight
+ 494: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.26.ssm_beta.weight
+ 495: 49152 | 4, 12288, 1, 1 | F32 | blk.26.ssm_conv1d.weight
+ 496: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_dt.bias
+ 497: 128 | 128, 1, 1, 1 | F32 | blk.26.ssm_norm.weight
+ 498: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.26.ssm_out.weight
+ 499: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.27.attn_k.weight
+ 500: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_k_norm.weight
+ 501: 3072 | 3072, 1, 1, 1 | F32 | blk.27.attn_norm.weight
+ 502: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.27.attn_output.weight
+ 503: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.27.attn_q.weight
+ 504: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_q_norm.weight
+ 505: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.27.attn_v.weight
+ 506: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.27.ffn_down_exps.weight
+ 507: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.27.ffn_down_shexp.weight
+ 508: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.27.ffn_gate_exps.weight
+ 509: 786432 | 3072, 256, 1, 1 | F32 | blk.27.ffn_gate_inp.weight
+ 510: 3072 | 3072, 1, 1, 1 | F32 | blk.27.ffn_gate_inp_shexp.weight
+ 511: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.27.ffn_gate_shexp.weight
+ 512: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.27.ffn_up_exps.weight
+ 513: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.27.ffn_up_shexp.weight
+ 514: 3072 | 3072, 1, 1, 1 | F32 | blk.27.post_attention_norm.weight
+ 515: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.28.attn_gate.weight
+ 516: 3072 | 3072, 1, 1, 1 | F32 | blk.28.attn_norm.weight
+ 517: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.28.attn_qkv.weight
+ 518: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.28.ffn_down_exps.weight
+ 519: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.28.ffn_down_shexp.weight
+ 520: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.28.ffn_gate_exps.weight
+ 521: 786432 | 3072, 256, 1, 1 | F32 | blk.28.ffn_gate_inp.weight
+ 522: 3072 | 3072, 1, 1, 1 | F32 | blk.28.ffn_gate_inp_shexp.weight
+ 523: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.28.ffn_gate_shexp.weight
+ 524: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.28.ffn_up_exps.weight
+ 525: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.28.ffn_up_shexp.weight
+ 526: 3072 | 3072, 1, 1, 1 | F32 | blk.28.post_attention_norm.weight
+ 527: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_a
+ 528: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.28.ssm_alpha.weight
+ 529: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.28.ssm_beta.weight
+ 530: 49152 | 4, 12288, 1, 1 | F32 | blk.28.ssm_conv1d.weight
+ 531: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_dt.bias
+ 532: 128 | 128, 1, 1, 1 | F32 | blk.28.ssm_norm.weight
+ 533: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.28.ssm_out.weight
+ 534: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.29.attn_gate.weight
+ 535: 3072 | 3072, 1, 1, 1 | F32 | blk.29.attn_norm.weight
+ 536: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.29.attn_qkv.weight
+ 537: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.29.ffn_down_exps.weight
+ 538: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.29.ffn_down_shexp.weight
+ 539: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.29.ffn_gate_exps.weight
+ 540: 786432 | 3072, 256, 1, 1 | F32 | blk.29.ffn_gate_inp.weight
+ 541: 3072 | 3072, 1, 1, 1 | F32 | blk.29.ffn_gate_inp_shexp.weight
+ 542: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.29.ffn_gate_shexp.weight
+ 543: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.29.ffn_up_exps.weight
+ 544: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.29.ffn_up_shexp.weight
+ 545: 3072 | 3072, 1, 1, 1 | F32 | blk.29.post_attention_norm.weight
+ 546: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_a
+ 547: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.29.ssm_alpha.weight
+ 548: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.29.ssm_beta.weight
+ 549: 49152 | 4, 12288, 1, 1 | F32 | blk.29.ssm_conv1d.weight
+ 550: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_dt.bias
+ 551: 128 | 128, 1, 1, 1 | F32 | blk.29.ssm_norm.weight
+ 552: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.29.ssm_out.weight
+ 553: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.30.attn_gate.weight
+ 554: 3072 | 3072, 1, 1, 1 | F32 | blk.30.attn_norm.weight
+ 555: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.30.attn_qkv.weight
+ 556: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.30.ffn_down_exps.weight
+ 557: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.30.ffn_down_shexp.weight
+ 558: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.30.ffn_gate_exps.weight
+ 559: 786432 | 3072, 256, 1, 1 | F32 | blk.30.ffn_gate_inp.weight
+ 560: 3072 | 3072, 1, 1, 1 | F32 | blk.30.ffn_gate_inp_shexp.weight
+ 561: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.30.ffn_gate_shexp.weight
+ 562: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.30.ffn_up_exps.weight
+ 563: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.30.ffn_up_shexp.weight
+ 564: 3072 | 3072, 1, 1, 1 | F32 | blk.30.post_attention_norm.weight
+ 565: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_a
+ 566: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.30.ssm_alpha.weight
+ 567: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.30.ssm_beta.weight
+ 568: 49152 | 4, 12288, 1, 1 | F32 | blk.30.ssm_conv1d.weight
+ 569: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_dt.bias
+ 570: 128 | 128, 1, 1, 1 | F32 | blk.30.ssm_norm.weight
+ 571: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.30.ssm_out.weight
+ 572: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.31.attn_k.weight
+ 573: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_k_norm.weight
+ 574: 3072 | 3072, 1, 1, 1 | F32 | blk.31.attn_norm.weight
+ 575: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.31.attn_output.weight
+ 576: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.31.attn_q.weight
+ 577: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_q_norm.weight
+ 578: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.31.attn_v.weight
+ 579: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.31.ffn_down_exps.weight
+ 580: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.31.ffn_down_shexp.weight
+ 581: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.31.ffn_gate_exps.weight
+ 582: 786432 | 3072, 256, 1, 1 | F32 | blk.31.ffn_gate_inp.weight
+ 583: 3072 | 3072, 1, 1, 1 | F32 | blk.31.ffn_gate_inp_shexp.weight
+ 584: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.31.ffn_gate_shexp.weight
+ 585: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.31.ffn_up_exps.weight
+ 586: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.31.ffn_up_shexp.weight
+ 587: 3072 | 3072, 1, 1, 1 | F32 | blk.31.post_attention_norm.weight
+ 588: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.32.attn_gate.weight
+ 589: 3072 | 3072, 1, 1, 1 | F32 | blk.32.attn_norm.weight
+ 590: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.32.attn_qkv.weight
+ 591: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.32.ffn_down_exps.weight
+ 592: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.32.ffn_down_shexp.weight
+ 593: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.32.ffn_gate_exps.weight
+ 594: 786432 | 3072, 256, 1, 1 | F32 | blk.32.ffn_gate_inp.weight
+ 595: 3072 | 3072, 1, 1, 1 | F32 | blk.32.ffn_gate_inp_shexp.weight
+ 596: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.32.ffn_gate_shexp.weight
+ 597: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.32.ffn_up_exps.weight
+ 598: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.32.ffn_up_shexp.weight
+ 599: 3072 | 3072, 1, 1, 1 | F32 | blk.32.post_attention_norm.weight
+ 600: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_a
+ 601: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.32.ssm_alpha.weight
+ 602: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.32.ssm_beta.weight
+ 603: 49152 | 4, 12288, 1, 1 | F32 | blk.32.ssm_conv1d.weight
+ 604: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_dt.bias
+ 605: 128 | 128, 1, 1, 1 | F32 | blk.32.ssm_norm.weight
+ 606: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.32.ssm_out.weight
+ 607: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.33.attn_gate.weight
+ 608: 3072 | 3072, 1, 1, 1 | F32 | blk.33.attn_norm.weight
+ 609: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.33.attn_qkv.weight
+ 610: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.33.ffn_down_exps.weight
+ 611: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.33.ffn_down_shexp.weight
+ 612: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.33.ffn_gate_exps.weight
+ 613: 786432 | 3072, 256, 1, 1 | F32 | blk.33.ffn_gate_inp.weight
+ 614: 3072 | 3072, 1, 1, 1 | F32 | blk.33.ffn_gate_inp_shexp.weight
+ 615: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.33.ffn_gate_shexp.weight
+ 616: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.33.ffn_up_exps.weight
+ 617: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.33.ffn_up_shexp.weight
+ 618: 3072 | 3072, 1, 1, 1 | F32 | blk.33.post_attention_norm.weight
+ 619: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_a
+ 620: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.33.ssm_alpha.weight
+ 621: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.33.ssm_beta.weight
+ 622: 49152 | 4, 12288, 1, 1 | F32 | blk.33.ssm_conv1d.weight
+ 623: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_dt.bias
+ 624: 128 | 128, 1, 1, 1 | F32 | blk.33.ssm_norm.weight
+ 625: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.33.ssm_out.weight
+ 626: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.34.attn_gate.weight
+ 627: 3072 | 3072, 1, 1, 1 | F32 | blk.34.attn_norm.weight
+ 628: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.34.attn_qkv.weight
+ 629: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.34.ffn_down_exps.weight
+ 630: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.34.ffn_down_shexp.weight
+ 631: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.34.ffn_gate_exps.weight
+ 632: 786432 | 3072, 256, 1, 1 | F32 | blk.34.ffn_gate_inp.weight
+ 633: 3072 | 3072, 1, 1, 1 | F32 | blk.34.ffn_gate_inp_shexp.weight
+ 634: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.34.ffn_gate_shexp.weight
+ 635: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.34.ffn_up_exps.weight
+ 636: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.34.ffn_up_shexp.weight
+ 637: 3072 | 3072, 1, 1, 1 | F32 | blk.34.post_attention_norm.weight
+ 638: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_a
+ 639: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.34.ssm_alpha.weight
+ 640: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.34.ssm_beta.weight
+ 641: 49152 | 4, 12288, 1, 1 | F32 | blk.34.ssm_conv1d.weight
+ 642: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_dt.bias
+ 643: 128 | 128, 1, 1, 1 | F32 | blk.34.ssm_norm.weight
+ 644: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.34.ssm_out.weight
+ 645: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.35.attn_k.weight
+ 646: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_k_norm.weight
+ 647: 3072 | 3072, 1, 1, 1 | F32 | blk.35.attn_norm.weight
+ 648: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.35.attn_output.weight
+ 649: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.35.attn_q.weight
+ 650: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_q_norm.weight
+ 651: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.35.attn_v.weight
+ 652: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.35.ffn_down_exps.weight
+ 653: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.35.ffn_down_shexp.weight
+ 654: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.35.ffn_gate_exps.weight
+ 655: 786432 | 3072, 256, 1, 1 | F32 | blk.35.ffn_gate_inp.weight
+ 656: 3072 | 3072, 1, 1, 1 | F32 | blk.35.ffn_gate_inp_shexp.weight
+ 657: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.35.ffn_gate_shexp.weight
+ 658: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.35.ffn_up_exps.weight
+ 659: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.35.ffn_up_shexp.weight
+ 660: 3072 | 3072, 1, 1, 1 | F32 | blk.35.post_attention_norm.weight
+ 661: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.36.attn_gate.weight
+ 662: 3072 | 3072, 1, 1, 1 | F32 | blk.36.attn_norm.weight
+ 663: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.36.attn_qkv.weight
+ 664: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.36.ffn_down_exps.weight
+ 665: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.36.ffn_down_shexp.weight
+ 666: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.36.ffn_gate_exps.weight
+ 667: 786432 | 3072, 256, 1, 1 | F32 | blk.36.ffn_gate_inp.weight
+ 668: 3072 | 3072, 1, 1, 1 | F32 | blk.36.ffn_gate_inp_shexp.weight
+ 669: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.36.ffn_gate_shexp.weight
+ 670: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.36.ffn_up_exps.weight
+ 671: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.36.ffn_up_shexp.weight
+ 672: 3072 | 3072, 1, 1, 1 | F32 | blk.36.post_attention_norm.weight
+ 673: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_a
+ 674: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.36.ssm_alpha.weight
+ 675: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.36.ssm_beta.weight
+ 676: 49152 | 4, 12288, 1, 1 | F32 | blk.36.ssm_conv1d.weight
+ 677: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_dt.bias
+ 678: 128 | 128, 1, 1, 1 | F32 | blk.36.ssm_norm.weight
+ 679: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.36.ssm_out.weight
+ 680: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.37.attn_gate.weight
+ 681: 3072 | 3072, 1, 1, 1 | F32 | blk.37.attn_norm.weight
+ 682: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.37.attn_qkv.weight
+ 683: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.37.ffn_down_exps.weight
+ 684: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.37.ffn_down_shexp.weight
+ 685: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.37.ffn_gate_exps.weight
+ 686: 786432 | 3072, 256, 1, 1 | F32 | blk.37.ffn_gate_inp.weight
+ 687: 3072 | 3072, 1, 1, 1 | F32 | blk.37.ffn_gate_inp_shexp.weight
+ 688: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.37.ffn_gate_shexp.weight
+ 689: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.37.ffn_up_exps.weight
+ 690: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.37.ffn_up_shexp.weight
+ 691: 3072 | 3072, 1, 1, 1 | F32 | blk.37.post_attention_norm.weight
+ 692: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_a
+ 693: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.37.ssm_alpha.weight
+ 694: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.37.ssm_beta.weight
+ 695: 49152 | 4, 12288, 1, 1 | F32 | blk.37.ssm_conv1d.weight
+ 696: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_dt.bias
+ 697: 128 | 128, 1, 1, 1 | F32 | blk.37.ssm_norm.weight
+ 698: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.37.ssm_out.weight
+ 699: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.38.attn_gate.weight
+ 700: 3072 | 3072, 1, 1, 1 | F32 | blk.38.attn_norm.weight
+ 701: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.38.attn_qkv.weight
+ 702: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.38.ffn_down_exps.weight
+ 703: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.38.ffn_down_shexp.weight
+ 704: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.38.ffn_gate_exps.weight
+ 705: 786432 | 3072, 256, 1, 1 | F32 | blk.38.ffn_gate_inp.weight
+ 706: 3072 | 3072, 1, 1, 1 | F32 | blk.38.ffn_gate_inp_shexp.weight
+ 707: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.38.ffn_gate_shexp.weight
+ 708: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.38.ffn_up_exps.weight
+ 709: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.38.ffn_up_shexp.weight
+ 710: 3072 | 3072, 1, 1, 1 | F32 | blk.38.post_attention_norm.weight
+ 711: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_a
+ 712: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.38.ssm_alpha.weight
+ 713: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.38.ssm_beta.weight
+ 714: 49152 | 4, 12288, 1, 1 | F32 | blk.38.ssm_conv1d.weight
+ 715: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_dt.bias
+ 716: 128 | 128, 1, 1, 1 | F32 | blk.38.ssm_norm.weight
+ 717: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.38.ssm_out.weight
+ 718: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.39.attn_k.weight
+ 719: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_k_norm.weight
+ 720: 3072 | 3072, 1, 1, 1 | F32 | blk.39.attn_norm.weight
+ 721: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.39.attn_output.weight
+ 722: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.39.attn_q.weight
+ 723: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_q_norm.weight
+ 724: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.39.attn_v.weight
+ 725: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.39.ffn_down_exps.weight
+ 726: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.39.ffn_down_shexp.weight
+ 727: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.39.ffn_gate_exps.weight
+ 728: 786432 | 3072, 256, 1, 1 | F32 | blk.39.ffn_gate_inp.weight
+ 729: 3072 | 3072, 1, 1, 1 | F32 | blk.39.ffn_gate_inp_shexp.weight
+ 730: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.39.ffn_gate_shexp.weight
+ 731: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.39.ffn_up_exps.weight
+ 732: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.39.ffn_up_shexp.weight
+ 733: 3072 | 3072, 1, 1, 1 | F32 | blk.39.post_attention_norm.weight
+ 734: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.40.attn_gate.weight
+ 735: 3072 | 3072, 1, 1, 1 | F32 | blk.40.attn_norm.weight
+ 736: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.40.attn_qkv.weight
+ 737: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.40.ffn_down_exps.weight
+ 738: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.40.ffn_down_shexp.weight
+ 739: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.40.ffn_gate_exps.weight
+ 740: 786432 | 3072, 256, 1, 1 | F32 | blk.40.ffn_gate_inp.weight
+ 741: 3072 | 3072, 1, 1, 1 | F32 | blk.40.ffn_gate_inp_shexp.weight
+ 742: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.40.ffn_gate_shexp.weight
+ 743: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.40.ffn_up_exps.weight
+ 744: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.40.ffn_up_shexp.weight
+ 745: 3072 | 3072, 1, 1, 1 | F32 | blk.40.post_attention_norm.weight
+ 746: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_a
+ 747: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.40.ssm_alpha.weight
+ 748: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.40.ssm_beta.weight
+ 749: 49152 | 4, 12288, 1, 1 | F32 | blk.40.ssm_conv1d.weight
+ 750: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_dt.bias
+ 751: 128 | 128, 1, 1, 1 | F32 | blk.40.ssm_norm.weight
+ 752: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.40.ssm_out.weight
+ 753: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.41.attn_gate.weight
+ 754: 3072 | 3072, 1, 1, 1 | F32 | blk.41.attn_norm.weight
+ 755: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.41.attn_qkv.weight
+ 756: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.41.ffn_down_exps.weight
+ 757: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.41.ffn_down_shexp.weight
+ 758: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.41.ffn_gate_exps.weight
+ 759: 786432 | 3072, 256, 1, 1 | F32 | blk.41.ffn_gate_inp.weight
+ 760: 3072 | 3072, 1, 1, 1 | F32 | blk.41.ffn_gate_inp_shexp.weight
+ 761: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.41.ffn_gate_shexp.weight
+ 762: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.41.ffn_up_exps.weight
+ 763: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.41.ffn_up_shexp.weight
+ 764: 3072 | 3072, 1, 1, 1 | F32 | blk.41.post_attention_norm.weight
+ 765: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_a
+ 766: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.41.ssm_alpha.weight
+ 767: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.41.ssm_beta.weight
+ 768: 49152 | 4, 12288, 1, 1 | F32 | blk.41.ssm_conv1d.weight
+ 769: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_dt.bias
+ 770: 128 | 128, 1, 1, 1 | F32 | blk.41.ssm_norm.weight
+ 771: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.41.ssm_out.weight
+ 772: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.42.attn_gate.weight
+ 773: 3072 | 3072, 1, 1, 1 | F32 | blk.42.attn_norm.weight
+ 774: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.42.attn_qkv.weight
+ 775: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.42.ffn_down_exps.weight
+ 776: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.42.ffn_down_shexp.weight
+ 777: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.42.ffn_gate_exps.weight
+ 778: 786432 | 3072, 256, 1, 1 | F32 | blk.42.ffn_gate_inp.weight
+ 779: 3072 | 3072, 1, 1, 1 | F32 | blk.42.ffn_gate_inp_shexp.weight
+ 780: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.42.ffn_gate_shexp.weight
+ 781: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.42.ffn_up_exps.weight
+ 782: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.42.ffn_up_shexp.weight
+ 783: 3072 | 3072, 1, 1, 1 | F32 | blk.42.post_attention_norm.weight
+ 784: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_a
+ 785: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.42.ssm_alpha.weight
+ 786: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.42.ssm_beta.weight
+ 787: 49152 | 4, 12288, 1, 1 | F32 | blk.42.ssm_conv1d.weight
+ 788: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_dt.bias
+ 789: 128 | 128, 1, 1, 1 | F32 | blk.42.ssm_norm.weight
+ 790: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.42.ssm_out.weight
+ 791: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.43.attn_k.weight
+ 792: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_k_norm.weight
+ 793: 3072 | 3072, 1, 1, 1 | F32 | blk.43.attn_norm.weight
+ 794: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.43.attn_output.weight
+ 795: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.43.attn_q.weight
+ 796: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_q_norm.weight
+ 797: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.43.attn_v.weight
+ 798: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.43.ffn_down_exps.weight
+ 799: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.43.ffn_down_shexp.weight
+ 800: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.43.ffn_gate_exps.weight
+ 801: 786432 | 3072, 256, 1, 1 | F32 | blk.43.ffn_gate_inp.weight
+ 802: 3072 | 3072, 1, 1, 1 | F32 | blk.43.ffn_gate_inp_shexp.weight
+ 803: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.43.ffn_gate_shexp.weight
+ 804: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.43.ffn_up_exps.weight
+ 805: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.43.ffn_up_shexp.weight
+ 806: 3072 | 3072, 1, 1, 1 | F32 | blk.43.post_attention_norm.weight
+ 807: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.44.attn_gate.weight
+ 808: 3072 | 3072, 1, 1, 1 | F32 | blk.44.attn_norm.weight
+ 809: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.44.attn_qkv.weight
+ 810: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.44.ffn_down_exps.weight
+ 811: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.44.ffn_down_shexp.weight
+ 812: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.44.ffn_gate_exps.weight
+ 813: 786432 | 3072, 256, 1, 1 | F32 | blk.44.ffn_gate_inp.weight
+ 814: 3072 | 3072, 1, 1, 1 | F32 | blk.44.ffn_gate_inp_shexp.weight
+ 815: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.44.ffn_gate_shexp.weight
+ 816: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.44.ffn_up_exps.weight
+ 817: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.44.ffn_up_shexp.weight
+ 818: 3072 | 3072, 1, 1, 1 | F32 | blk.44.post_attention_norm.weight
+ 819: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_a
+ 820: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.44.ssm_alpha.weight
+ 821: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.44.ssm_beta.weight
+ 822: 49152 | 4, 12288, 1, 1 | F32 | blk.44.ssm_conv1d.weight
+ 823: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_dt.bias
+ 824: 128 | 128, 1, 1, 1 | F32 | blk.44.ssm_norm.weight
+ 825: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.44.ssm_out.weight
+ 826: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.45.attn_gate.weight
+ 827: 3072 | 3072, 1, 1, 1 | F32 | blk.45.attn_norm.weight
+ 828: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.45.attn_qkv.weight
+ 829: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.45.ffn_down_exps.weight
+ 830: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.45.ffn_down_shexp.weight
+ 831: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.45.ffn_gate_exps.weight
+ 832: 786432 | 3072, 256, 1, 1 | F32 | blk.45.ffn_gate_inp.weight
+ 833: 3072 | 3072, 1, 1, 1 | F32 | blk.45.ffn_gate_inp_shexp.weight
+ 834: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.45.ffn_gate_shexp.weight
+ 835: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.45.ffn_up_exps.weight
+ 836: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.45.ffn_up_shexp.weight
+ 837: 3072 | 3072, 1, 1, 1 | F32 | blk.45.post_attention_norm.weight
+ 838: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_a
+ 839: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.45.ssm_alpha.weight
+ 840: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.45.ssm_beta.weight
+ 841: 49152 | 4, 12288, 1, 1 | F32 | blk.45.ssm_conv1d.weight
+ 842: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_dt.bias
+ 843: 128 | 128, 1, 1, 1 | F32 | blk.45.ssm_norm.weight
+ 844: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.45.ssm_out.weight
+ 845: 25165824 | 3072, 8192, 1, 1 | Q8_0 | blk.46.attn_gate.weight
+ 846: 3072 | 3072, 1, 1, 1 | F32 | blk.46.attn_norm.weight
+ 847: 37748736 | 3072, 12288, 1, 1 | Q8_0 | blk.46.attn_qkv.weight
+ 848: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.46.ffn_down_exps.weight
+ 849: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.46.ffn_down_shexp.weight
+ 850: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.46.ffn_gate_exps.weight
+ 851: 786432 | 3072, 256, 1, 1 | F32 | blk.46.ffn_gate_inp.weight
+ 852: 3072 | 3072, 1, 1, 1 | F32 | blk.46.ffn_gate_inp_shexp.weight
+ 853: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.46.ffn_gate_shexp.weight
+ 854: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.46.ffn_up_exps.weight
+ 855: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.46.ffn_up_shexp.weight
+ 856: 3072 | 3072, 1, 1, 1 | F32 | blk.46.post_attention_norm.weight
+ 857: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_a
+ 858: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.46.ssm_alpha.weight
+ 859: 196608 | 3072, 64, 1, 1 | Q8_0 | blk.46.ssm_beta.weight
+ 860: 49152 | 4, 12288, 1, 1 | F32 | blk.46.ssm_conv1d.weight
+ 861: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_dt.bias
+ 862: 128 | 128, 1, 1, 1 | F32 | blk.46.ssm_norm.weight
+ 863: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.46.ssm_out.weight
+ 864: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.47.attn_k.weight
+ 865: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_k_norm.weight
+ 866: 3072 | 3072, 1, 1, 1 | F32 | blk.47.attn_norm.weight
+ 867: 25165824 | 8192, 3072, 1, 1 | Q8_0 | blk.47.attn_output.weight
+ 868: 50331648 | 3072, 16384, 1, 1 | Q8_0 | blk.47.attn_q.weight
+ 869: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_q_norm.weight
+ 870: 1572864 | 3072, 512, 1, 1 | Q8_0 | blk.47.attn_v.weight
+ 871: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.47.ffn_down_exps.weight
+ 872: 3145728 | 1024, 3072, 1, 1 | Q8_0 | blk.47.ffn_down_shexp.weight
+ 873: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.47.ffn_gate_exps.weight
+ 874: 786432 | 3072, 256, 1, 1 | F32 | blk.47.ffn_gate_inp.weight
+ 875: 3072 | 3072, 1, 1, 1 | F32 | blk.47.ffn_gate_inp_shexp.weight
+ 876: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.47.ffn_gate_shexp.weight
+ 877: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.47.ffn_up_exps.weight
+ 878: 3145728 | 3072, 1024, 1, 1 | Q8_0 | blk.47.ffn_up_shexp.weight
+ 879: 3072 | 3072, 1, 1, 1 | F32 | blk.47.post_attention_norm.weight
+
+
+
+**MQ516_special (Mixed Quant Q5_K - BF16, Special Purpose)**
+
+gguf-dump
+
+ --leave-output-tensor
+ --tensor-type "blk\.(0|1|2|4|5|6|8|9|10)\..*ffn_(down|gate|up)_exps\.weight=q5_k"
+ --tensor-type ".*ffn_(down|gate|up)_shexp\.weight=bf16"
+ --tensor-type "blk\.(3|7|11|15|19|23|27|31|35|39|43|47)\..*attn_(q|k|v|output)\.weight=bf16"
+ --tensor-type ".*ssm_(alpha|beta|out)\.weight=bf16"
+ --tensor-type "token_embd\.weight=bf16"
+ --tensor-type ".*attn_(qkv|gate)\.weight=bf16"
+ Q6_K
+
+ 1: UINT32 | 1 | GGUF.version = 3
+ 2: UINT64 | 1 | GGUF.tensor_count = 879
+ 3: UINT64 | 1 | GGUF.kv_count = 43
+ 4: STRING | 1 | general.architecture = 'qwen35moe'
+ 5: STRING | 1 | general.type = 'model'
+ 6: INT32 | 1 | general.sampling.top_k = 20
+ 7: FLOAT32 | 1 | general.sampling.top_p = 0.949999988079071
+ 8: FLOAT32 | 1 | general.sampling.temp = 0.6000000238418579
+ 9: STRING | 1 | general.name = 'Qwen3.5 122B A10B Wangzhang Abliterix No MTP'
+ 10: STRING | 1 | general.size_label = '256x6.6B'
+ 11: UINT32 | 1 | general.base_model.count = 1
+ 12: STRING | 1 | general.base_model.0.name = 'Qwen3.5 122B A10B'
+ 13: STRING | 1 | general.base_model.0.organization = 'Qwen'
+ 14: STRING | 1 | general.base_model.0.repo_url = 'https://huggingface.co/Qwen/Qwen3.5-122B-A10B'
+ 15: [STRING] | 5 | general.tags = ['abliterix', 'uncensored', 'decensored', 'abliterated', 'text-generation']
+ 16: UINT32 | 1 | qwen35moe.block_count = 48
+ 17: UINT32 | 1 | qwen35moe.context_length = 262144
+ 18: UINT32 | 1 | qwen35moe.embedding_length = 3072
+ 19: UINT32 | 1 | qwen35moe.attention.head_count = 32
+ 20: UINT32 | 1 | qwen35moe.attention.head_count_kv = 2
+ 21: [INT32] | 4 | qwen35moe.rope.dimension_sections = [11, 11, 10, 0]
+ 22: FLOAT32 | 1 | qwen35moe.rope.freq_base = 10000000.0
+ 23: FLOAT32 | 1 | qwen35moe.attention.layer_norm_rms_epsilon = 9.999999974752427e-07
+ 24: UINT32 | 1 | qwen35moe.expert_count = 256
+ 25: UINT32 | 1 | qwen35moe.expert_used_count = 8
+ 26: UINT32 | 1 | qwen35moe.attention.key_length = 256
+ 27: UINT32 | 1 | qwen35moe.attention.value_length = 256
+ 28: UINT32 | 1 | qwen35moe.expert_feed_forward_length = 1024
+ 29: UINT32 | 1 | qwen35moe.expert_shared_feed_forward_length = 1024
+ 30: UINT32 | 1 | qwen35moe.ssm.conv_kernel = 4
+ 31: UINT32 | 1 | qwen35moe.ssm.state_size = 128
+ 32: UINT32 | 1 | qwen35moe.ssm.group_count = 16
+ 33: UINT32 | 1 | qwen35moe.ssm.time_step_rank = 64
+ 34: UINT32 | 1 | qwen35moe.ssm.inner_size = 8192
+ 35: UINT32 | 1 | qwen35moe.full_attention_interval = 4
+ 36: UINT32 | 1 | qwen35moe.rope.dimension_count = 64
+ 37: STRING | 1 | tokenizer.ggml.model = 'gpt2'
+ 38: STRING | 1 | tokenizer.ggml.pre = 'qwen35'
+ 39: [STRING] | 248320 | tokenizer.ggml.tokens = ['!', '"', '#', '$', '%', '&', ...]
+ 40: [INT32] | 248320 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...]
+ 41: [STRING] | 247587 | tokenizer.ggml.merges = ['Ġ Ġ', 'ĠĠ ĠĠ', 'i n', 'Ġ t', 'ĠĠĠĠ ĠĠĠĠ', 'e r', ...]
+ 42: UINT32 | 1 | tokenizer.ggml.eos_token_id = 248046
+ 43: UINT32 | 1 | tokenizer.ggml.padding_token_id = 248044
+ 44: STRING | 1 | tokenizer.chat_template = '{%- set image_count = namespace(value=0) %}\n{%- set video...'
+ 45: UINT32 | 1 | general.quantization_version = 2
+ 46: UINT32 | 1 | general.file_type = 18
+
+ 1: 762839040 | 3072, 248320, 1, 1 | BF16 | output.weight
+ 2: 3072 | 3072, 1, 1, 1 | F32 | output_norm.weight
+ 3: 762839040 | 3072, 248320, 1, 1 | BF16 | token_embd.weight
+ 4: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.0.attn_gate.weight
+ 5: 3072 | 3072, 1, 1, 1 | F32 | blk.0.attn_norm.weight
+ 6: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.0.attn_qkv.weight
+ 7: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.0.ffn_down_exps.weight
+ 8: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.0.ffn_down_shexp.weight
+ 9: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.0.ffn_gate_exps.weight
+ 10: 786432 | 3072, 256, 1, 1 | F32 | blk.0.ffn_gate_inp.weight
+ 11: 3072 | 3072, 1, 1, 1 | F32 | blk.0.ffn_gate_inp_shexp.weight
+ 12: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_gate_shexp.weight
+ 13: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.0.ffn_up_exps.weight
+ 14: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_up_shexp.weight
+ 15: 3072 | 3072, 1, 1, 1 | F32 | blk.0.post_attention_norm.weight
+ 16: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_a
+ 17: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_alpha.weight
+ 18: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_beta.weight
+ 19: 49152 | 4, 12288, 1, 1 | F32 | blk.0.ssm_conv1d.weight
+ 20: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_dt.bias
+ 21: 128 | 128, 1, 1, 1 | F32 | blk.0.ssm_norm.weight
+ 22: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.0.ssm_out.weight
+ 23: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.1.attn_gate.weight
+ 24: 3072 | 3072, 1, 1, 1 | F32 | blk.1.attn_norm.weight
+ 25: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.1.attn_qkv.weight
+ 26: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.1.ffn_down_exps.weight
+ 27: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.1.ffn_down_shexp.weight
+ 28: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.1.ffn_gate_exps.weight
+ 29: 786432 | 3072, 256, 1, 1 | F32 | blk.1.ffn_gate_inp.weight
+ 30: 3072 | 3072, 1, 1, 1 | F32 | blk.1.ffn_gate_inp_shexp.weight
+ 31: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_gate_shexp.weight
+ 32: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.1.ffn_up_exps.weight
+ 33: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_up_shexp.weight
+ 34: 3072 | 3072, 1, 1, 1 | F32 | blk.1.post_attention_norm.weight
+ 35: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_a
+ 36: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_alpha.weight
+ 37: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_beta.weight
+ 38: 49152 | 4, 12288, 1, 1 | F32 | blk.1.ssm_conv1d.weight
+ 39: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_dt.bias
+ 40: 128 | 128, 1, 1, 1 | F32 | blk.1.ssm_norm.weight
+ 41: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.1.ssm_out.weight
+ 42: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.2.attn_gate.weight
+ 43: 3072 | 3072, 1, 1, 1 | F32 | blk.2.attn_norm.weight
+ 44: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.2.attn_qkv.weight
+ 45: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.2.ffn_down_exps.weight
+ 46: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.2.ffn_down_shexp.weight
+ 47: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.2.ffn_gate_exps.weight
+ 48: 786432 | 3072, 256, 1, 1 | F32 | blk.2.ffn_gate_inp.weight
+ 49: 3072 | 3072, 1, 1, 1 | F32 | blk.2.ffn_gate_inp_shexp.weight
+ 50: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_gate_shexp.weight
+ 51: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.2.ffn_up_exps.weight
+ 52: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_up_shexp.weight
+ 53: 3072 | 3072, 1, 1, 1 | F32 | blk.2.post_attention_norm.weight
+ 54: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_a
+ 55: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_alpha.weight
+ 56: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_beta.weight
+ 57: 49152 | 4, 12288, 1, 1 | F32 | blk.2.ssm_conv1d.weight
+ 58: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_dt.bias
+ 59: 128 | 128, 1, 1, 1 | F32 | blk.2.ssm_norm.weight
+ 60: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.2.ssm_out.weight
+ 61: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_k.weight
+ 62: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_k_norm.weight
+ 63: 3072 | 3072, 1, 1, 1 | F32 | blk.3.attn_norm.weight
+ 64: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.3.attn_output.weight
+ 65: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.3.attn_q.weight
+ 66: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_q_norm.weight
+ 67: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_v.weight
+ 68: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.3.ffn_down_exps.weight
+ 69: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.3.ffn_down_shexp.weight
+ 70: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.3.ffn_gate_exps.weight
+ 71: 786432 | 3072, 256, 1, 1 | F32 | blk.3.ffn_gate_inp.weight
+ 72: 3072 | 3072, 1, 1, 1 | F32 | blk.3.ffn_gate_inp_shexp.weight
+ 73: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_gate_shexp.weight
+ 74: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.3.ffn_up_exps.weight
+ 75: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_up_shexp.weight
+ 76: 3072 | 3072, 1, 1, 1 | F32 | blk.3.post_attention_norm.weight
+ 77: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.4.attn_gate.weight
+ 78: 3072 | 3072, 1, 1, 1 | F32 | blk.4.attn_norm.weight
+ 79: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.4.attn_qkv.weight
+ 80: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.4.ffn_down_exps.weight
+ 81: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.4.ffn_down_shexp.weight
+ 82: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.4.ffn_gate_exps.weight
+ 83: 786432 | 3072, 256, 1, 1 | F32 | blk.4.ffn_gate_inp.weight
+ 84: 3072 | 3072, 1, 1, 1 | F32 | blk.4.ffn_gate_inp_shexp.weight
+ 85: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_gate_shexp.weight
+ 86: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.4.ffn_up_exps.weight
+ 87: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_up_shexp.weight
+ 88: 3072 | 3072, 1, 1, 1 | F32 | blk.4.post_attention_norm.weight
+ 89: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_a
+ 90: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_alpha.weight
+ 91: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_beta.weight
+ 92: 49152 | 4, 12288, 1, 1 | F32 | blk.4.ssm_conv1d.weight
+ 93: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_dt.bias
+ 94: 128 | 128, 1, 1, 1 | F32 | blk.4.ssm_norm.weight
+ 95: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.4.ssm_out.weight
+ 96: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.5.attn_gate.weight
+ 97: 3072 | 3072, 1, 1, 1 | F32 | blk.5.attn_norm.weight
+ 98: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.5.attn_qkv.weight
+ 99: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.5.ffn_down_exps.weight
+ 100: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.5.ffn_down_shexp.weight
+ 101: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.5.ffn_gate_exps.weight
+ 102: 786432 | 3072, 256, 1, 1 | F32 | blk.5.ffn_gate_inp.weight
+ 103: 3072 | 3072, 1, 1, 1 | F32 | blk.5.ffn_gate_inp_shexp.weight
+ 104: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_gate_shexp.weight
+ 105: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.5.ffn_up_exps.weight
+ 106: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_up_shexp.weight
+ 107: 3072 | 3072, 1, 1, 1 | F32 | blk.5.post_attention_norm.weight
+ 108: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_a
+ 109: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_alpha.weight
+ 110: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_beta.weight
+ 111: 49152 | 4, 12288, 1, 1 | F32 | blk.5.ssm_conv1d.weight
+ 112: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_dt.bias
+ 113: 128 | 128, 1, 1, 1 | F32 | blk.5.ssm_norm.weight
+ 114: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.5.ssm_out.weight
+ 115: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.6.attn_gate.weight
+ 116: 3072 | 3072, 1, 1, 1 | F32 | blk.6.attn_norm.weight
+ 117: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.6.attn_qkv.weight
+ 118: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.6.ffn_down_exps.weight
+ 119: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.6.ffn_down_shexp.weight
+ 120: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.6.ffn_gate_exps.weight
+ 121: 786432 | 3072, 256, 1, 1 | F32 | blk.6.ffn_gate_inp.weight
+ 122: 3072 | 3072, 1, 1, 1 | F32 | blk.6.ffn_gate_inp_shexp.weight
+ 123: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_gate_shexp.weight
+ 124: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.6.ffn_up_exps.weight
+ 125: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_up_shexp.weight
+ 126: 3072 | 3072, 1, 1, 1 | F32 | blk.6.post_attention_norm.weight
+ 127: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_a
+ 128: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_alpha.weight
+ 129: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_beta.weight
+ 130: 49152 | 4, 12288, 1, 1 | F32 | blk.6.ssm_conv1d.weight
+ 131: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_dt.bias
+ 132: 128 | 128, 1, 1, 1 | F32 | blk.6.ssm_norm.weight
+ 133: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.6.ssm_out.weight
+ 134: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_k.weight
+ 135: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_k_norm.weight
+ 136: 3072 | 3072, 1, 1, 1 | F32 | blk.7.attn_norm.weight
+ 137: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.7.attn_output.weight
+ 138: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.7.attn_q.weight
+ 139: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_q_norm.weight
+ 140: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_v.weight
+ 141: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.7.ffn_down_exps.weight
+ 142: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.7.ffn_down_shexp.weight
+ 143: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.7.ffn_gate_exps.weight
+ 144: 786432 | 3072, 256, 1, 1 | F32 | blk.7.ffn_gate_inp.weight
+ 145: 3072 | 3072, 1, 1, 1 | F32 | blk.7.ffn_gate_inp_shexp.weight
+ 146: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_gate_shexp.weight
+ 147: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.7.ffn_up_exps.weight
+ 148: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_up_shexp.weight
+ 149: 3072 | 3072, 1, 1, 1 | F32 | blk.7.post_attention_norm.weight
+ 150: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.8.attn_gate.weight
+ 151: 3072 | 3072, 1, 1, 1 | F32 | blk.8.attn_norm.weight
+ 152: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.8.attn_qkv.weight
+ 153: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.8.ffn_down_exps.weight
+ 154: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.8.ffn_down_shexp.weight
+ 155: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.8.ffn_gate_exps.weight
+ 156: 786432 | 3072, 256, 1, 1 | F32 | blk.8.ffn_gate_inp.weight
+ 157: 3072 | 3072, 1, 1, 1 | F32 | blk.8.ffn_gate_inp_shexp.weight
+ 158: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_gate_shexp.weight
+ 159: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.8.ffn_up_exps.weight
+ 160: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_up_shexp.weight
+ 161: 3072 | 3072, 1, 1, 1 | F32 | blk.8.post_attention_norm.weight
+ 162: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_a
+ 163: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_alpha.weight
+ 164: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_beta.weight
+ 165: 49152 | 4, 12288, 1, 1 | F32 | blk.8.ssm_conv1d.weight
+ 166: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_dt.bias
+ 167: 128 | 128, 1, 1, 1 | F32 | blk.8.ssm_norm.weight
+ 168: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.8.ssm_out.weight
+ 169: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.9.attn_gate.weight
+ 170: 3072 | 3072, 1, 1, 1 | F32 | blk.9.attn_norm.weight
+ 171: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.9.attn_qkv.weight
+ 172: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.9.ffn_down_exps.weight
+ 173: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.9.ffn_down_shexp.weight
+ 174: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.9.ffn_gate_exps.weight
+ 175: 786432 | 3072, 256, 1, 1 | F32 | blk.9.ffn_gate_inp.weight
+ 176: 3072 | 3072, 1, 1, 1 | F32 | blk.9.ffn_gate_inp_shexp.weight
+ 177: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_gate_shexp.weight
+ 178: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.9.ffn_up_exps.weight
+ 179: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_up_shexp.weight
+ 180: 3072 | 3072, 1, 1, 1 | F32 | blk.9.post_attention_norm.weight
+ 181: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_a
+ 182: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_alpha.weight
+ 183: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_beta.weight
+ 184: 49152 | 4, 12288, 1, 1 | F32 | blk.9.ssm_conv1d.weight
+ 185: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_dt.bias
+ 186: 128 | 128, 1, 1, 1 | F32 | blk.9.ssm_norm.weight
+ 187: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.9.ssm_out.weight
+ 188: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.10.attn_gate.weight
+ 189: 3072 | 3072, 1, 1, 1 | F32 | blk.10.attn_norm.weight
+ 190: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.10.attn_qkv.weight
+ 191: 805306368 | 1024, 3072, 256, 1 | Q5_K | blk.10.ffn_down_exps.weight
+ 192: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.10.ffn_down_shexp.weight
+ 193: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.10.ffn_gate_exps.weight
+ 194: 786432 | 3072, 256, 1, 1 | F32 | blk.10.ffn_gate_inp.weight
+ 195: 3072 | 3072, 1, 1, 1 | F32 | blk.10.ffn_gate_inp_shexp.weight
+ 196: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_gate_shexp.weight
+ 197: 805306368 | 3072, 1024, 256, 1 | Q5_K | blk.10.ffn_up_exps.weight
+ 198: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_up_shexp.weight
+ 199: 3072 | 3072, 1, 1, 1 | F32 | blk.10.post_attention_norm.weight
+ 200: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_a
+ 201: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_alpha.weight
+ 202: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_beta.weight
+ 203: 49152 | 4, 12288, 1, 1 | F32 | blk.10.ssm_conv1d.weight
+ 204: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_dt.bias
+ 205: 128 | 128, 1, 1, 1 | F32 | blk.10.ssm_norm.weight
+ 206: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.10.ssm_out.weight
+ 207: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_k.weight
+ 208: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_k_norm.weight
+ 209: 3072 | 3072, 1, 1, 1 | F32 | blk.11.attn_norm.weight
+ 210: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.11.attn_output.weight
+ 211: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.11.attn_q.weight
+ 212: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_q_norm.weight
+ 213: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_v.weight
+ 214: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.11.ffn_down_exps.weight
+ 215: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.11.ffn_down_shexp.weight
+ 216: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.11.ffn_gate_exps.weight
+ 217: 786432 | 3072, 256, 1, 1 | F32 | blk.11.ffn_gate_inp.weight
+ 218: 3072 | 3072, 1, 1, 1 | F32 | blk.11.ffn_gate_inp_shexp.weight
+ 219: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_gate_shexp.weight
+ 220: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.11.ffn_up_exps.weight
+ 221: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_up_shexp.weight
+ 222: 3072 | 3072, 1, 1, 1 | F32 | blk.11.post_attention_norm.weight
+ 223: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.12.attn_gate.weight
+ 224: 3072 | 3072, 1, 1, 1 | F32 | blk.12.attn_norm.weight
+ 225: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.12.attn_qkv.weight
+ 226: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.12.ffn_down_exps.weight
+ 227: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.12.ffn_down_shexp.weight
+ 228: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.12.ffn_gate_exps.weight
+ 229: 786432 | 3072, 256, 1, 1 | F32 | blk.12.ffn_gate_inp.weight
+ 230: 3072 | 3072, 1, 1, 1 | F32 | blk.12.ffn_gate_inp_shexp.weight
+ 231: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_gate_shexp.weight
+ 232: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.12.ffn_up_exps.weight
+ 233: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_up_shexp.weight
+ 234: 3072 | 3072, 1, 1, 1 | F32 | blk.12.post_attention_norm.weight
+ 235: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_a
+ 236: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_alpha.weight
+ 237: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_beta.weight
+ 238: 49152 | 4, 12288, 1, 1 | F32 | blk.12.ssm_conv1d.weight
+ 239: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_dt.bias
+ 240: 128 | 128, 1, 1, 1 | F32 | blk.12.ssm_norm.weight
+ 241: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.12.ssm_out.weight
+ 242: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.13.attn_gate.weight
+ 243: 3072 | 3072, 1, 1, 1 | F32 | blk.13.attn_norm.weight
+ 244: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.13.attn_qkv.weight
+ 245: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.13.ffn_down_exps.weight
+ 246: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.13.ffn_down_shexp.weight
+ 247: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.13.ffn_gate_exps.weight
+ 248: 786432 | 3072, 256, 1, 1 | F32 | blk.13.ffn_gate_inp.weight
+ 249: 3072 | 3072, 1, 1, 1 | F32 | blk.13.ffn_gate_inp_shexp.weight
+ 250: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_gate_shexp.weight
+ 251: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.13.ffn_up_exps.weight
+ 252: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_up_shexp.weight
+ 253: 3072 | 3072, 1, 1, 1 | F32 | blk.13.post_attention_norm.weight
+ 254: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_a
+ 255: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_alpha.weight
+ 256: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_beta.weight
+ 257: 49152 | 4, 12288, 1, 1 | F32 | blk.13.ssm_conv1d.weight
+ 258: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_dt.bias
+ 259: 128 | 128, 1, 1, 1 | F32 | blk.13.ssm_norm.weight
+ 260: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.13.ssm_out.weight
+ 261: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.14.attn_gate.weight
+ 262: 3072 | 3072, 1, 1, 1 | F32 | blk.14.attn_norm.weight
+ 263: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.14.attn_qkv.weight
+ 264: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.14.ffn_down_exps.weight
+ 265: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.14.ffn_down_shexp.weight
+ 266: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.14.ffn_gate_exps.weight
+ 267: 786432 | 3072, 256, 1, 1 | F32 | blk.14.ffn_gate_inp.weight
+ 268: 3072 | 3072, 1, 1, 1 | F32 | blk.14.ffn_gate_inp_shexp.weight
+ 269: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_gate_shexp.weight
+ 270: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.14.ffn_up_exps.weight
+ 271: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_up_shexp.weight
+ 272: 3072 | 3072, 1, 1, 1 | F32 | blk.14.post_attention_norm.weight
+ 273: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_a
+ 274: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_alpha.weight
+ 275: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_beta.weight
+ 276: 49152 | 4, 12288, 1, 1 | F32 | blk.14.ssm_conv1d.weight
+ 277: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_dt.bias
+ 278: 128 | 128, 1, 1, 1 | F32 | blk.14.ssm_norm.weight
+ 279: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.14.ssm_out.weight
+ 280: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_k.weight
+ 281: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_k_norm.weight
+ 282: 3072 | 3072, 1, 1, 1 | F32 | blk.15.attn_norm.weight
+ 283: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.15.attn_output.weight
+ 284: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.15.attn_q.weight
+ 285: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_q_norm.weight
+ 286: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_v.weight
+ 287: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.15.ffn_down_exps.weight
+ 288: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.15.ffn_down_shexp.weight
+ 289: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.15.ffn_gate_exps.weight
+ 290: 786432 | 3072, 256, 1, 1 | F32 | blk.15.ffn_gate_inp.weight
+ 291: 3072 | 3072, 1, 1, 1 | F32 | blk.15.ffn_gate_inp_shexp.weight
+ 292: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_gate_shexp.weight
+ 293: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.15.ffn_up_exps.weight
+ 294: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_up_shexp.weight
+ 295: 3072 | 3072, 1, 1, 1 | F32 | blk.15.post_attention_norm.weight
+ 296: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.16.attn_gate.weight
+ 297: 3072 | 3072, 1, 1, 1 | F32 | blk.16.attn_norm.weight
+ 298: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.16.attn_qkv.weight
+ 299: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.16.ffn_down_exps.weight
+ 300: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.16.ffn_down_shexp.weight
+ 301: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.16.ffn_gate_exps.weight
+ 302: 786432 | 3072, 256, 1, 1 | F32 | blk.16.ffn_gate_inp.weight
+ 303: 3072 | 3072, 1, 1, 1 | F32 | blk.16.ffn_gate_inp_shexp.weight
+ 304: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_gate_shexp.weight
+ 305: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.16.ffn_up_exps.weight
+ 306: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_up_shexp.weight
+ 307: 3072 | 3072, 1, 1, 1 | F32 | blk.16.post_attention_norm.weight
+ 308: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_a
+ 309: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_alpha.weight
+ 310: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_beta.weight
+ 311: 49152 | 4, 12288, 1, 1 | F32 | blk.16.ssm_conv1d.weight
+ 312: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_dt.bias
+ 313: 128 | 128, 1, 1, 1 | F32 | blk.16.ssm_norm.weight
+ 314: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.16.ssm_out.weight
+ 315: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.17.attn_gate.weight
+ 316: 3072 | 3072, 1, 1, 1 | F32 | blk.17.attn_norm.weight
+ 317: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.17.attn_qkv.weight
+ 318: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.17.ffn_down_exps.weight
+ 319: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.17.ffn_down_shexp.weight
+ 320: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.17.ffn_gate_exps.weight
+ 321: 786432 | 3072, 256, 1, 1 | F32 | blk.17.ffn_gate_inp.weight
+ 322: 3072 | 3072, 1, 1, 1 | F32 | blk.17.ffn_gate_inp_shexp.weight
+ 323: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_gate_shexp.weight
+ 324: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.17.ffn_up_exps.weight
+ 325: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_up_shexp.weight
+ 326: 3072 | 3072, 1, 1, 1 | F32 | blk.17.post_attention_norm.weight
+ 327: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_a
+ 328: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_alpha.weight
+ 329: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_beta.weight
+ 330: 49152 | 4, 12288, 1, 1 | F32 | blk.17.ssm_conv1d.weight
+ 331: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_dt.bias
+ 332: 128 | 128, 1, 1, 1 | F32 | blk.17.ssm_norm.weight
+ 333: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.17.ssm_out.weight
+ 334: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.18.attn_gate.weight
+ 335: 3072 | 3072, 1, 1, 1 | F32 | blk.18.attn_norm.weight
+ 336: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.18.attn_qkv.weight
+ 337: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.18.ffn_down_exps.weight
+ 338: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.18.ffn_down_shexp.weight
+ 339: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.18.ffn_gate_exps.weight
+ 340: 786432 | 3072, 256, 1, 1 | F32 | blk.18.ffn_gate_inp.weight
+ 341: 3072 | 3072, 1, 1, 1 | F32 | blk.18.ffn_gate_inp_shexp.weight
+ 342: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_gate_shexp.weight
+ 343: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.18.ffn_up_exps.weight
+ 344: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_up_shexp.weight
+ 345: 3072 | 3072, 1, 1, 1 | F32 | blk.18.post_attention_norm.weight
+ 346: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_a
+ 347: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_alpha.weight
+ 348: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_beta.weight
+ 349: 49152 | 4, 12288, 1, 1 | F32 | blk.18.ssm_conv1d.weight
+ 350: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_dt.bias
+ 351: 128 | 128, 1, 1, 1 | F32 | blk.18.ssm_norm.weight
+ 352: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.18.ssm_out.weight
+ 353: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_k.weight
+ 354: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_k_norm.weight
+ 355: 3072 | 3072, 1, 1, 1 | F32 | blk.19.attn_norm.weight
+ 356: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.19.attn_output.weight
+ 357: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.19.attn_q.weight
+ 358: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_q_norm.weight
+ 359: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_v.weight
+ 360: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.19.ffn_down_exps.weight
+ 361: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.19.ffn_down_shexp.weight
+ 362: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.19.ffn_gate_exps.weight
+ 363: 786432 | 3072, 256, 1, 1 | F32 | blk.19.ffn_gate_inp.weight
+ 364: 3072 | 3072, 1, 1, 1 | F32 | blk.19.ffn_gate_inp_shexp.weight
+ 365: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_gate_shexp.weight
+ 366: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.19.ffn_up_exps.weight
+ 367: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_up_shexp.weight
+ 368: 3072 | 3072, 1, 1, 1 | F32 | blk.19.post_attention_norm.weight
+ 369: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.20.attn_gate.weight
+ 370: 3072 | 3072, 1, 1, 1 | F32 | blk.20.attn_norm.weight
+ 371: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.20.attn_qkv.weight
+ 372: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.20.ffn_down_exps.weight
+ 373: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.20.ffn_down_shexp.weight
+ 374: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.20.ffn_gate_exps.weight
+ 375: 786432 | 3072, 256, 1, 1 | F32 | blk.20.ffn_gate_inp.weight
+ 376: 3072 | 3072, 1, 1, 1 | F32 | blk.20.ffn_gate_inp_shexp.weight
+ 377: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_gate_shexp.weight
+ 378: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.20.ffn_up_exps.weight
+ 379: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_up_shexp.weight
+ 380: 3072 | 3072, 1, 1, 1 | F32 | blk.20.post_attention_norm.weight
+ 381: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_a
+ 382: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_alpha.weight
+ 383: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_beta.weight
+ 384: 49152 | 4, 12288, 1, 1 | F32 | blk.20.ssm_conv1d.weight
+ 385: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_dt.bias
+ 386: 128 | 128, 1, 1, 1 | F32 | blk.20.ssm_norm.weight
+ 387: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.20.ssm_out.weight
+ 388: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.21.attn_gate.weight
+ 389: 3072 | 3072, 1, 1, 1 | F32 | blk.21.attn_norm.weight
+ 390: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.21.attn_qkv.weight
+ 391: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.21.ffn_down_exps.weight
+ 392: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.21.ffn_down_shexp.weight
+ 393: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.21.ffn_gate_exps.weight
+ 394: 786432 | 3072, 256, 1, 1 | F32 | blk.21.ffn_gate_inp.weight
+ 395: 3072 | 3072, 1, 1, 1 | F32 | blk.21.ffn_gate_inp_shexp.weight
+ 396: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_gate_shexp.weight
+ 397: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.21.ffn_up_exps.weight
+ 398: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_up_shexp.weight
+ 399: 3072 | 3072, 1, 1, 1 | F32 | blk.21.post_attention_norm.weight
+ 400: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_a
+ 401: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_alpha.weight
+ 402: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_beta.weight
+ 403: 49152 | 4, 12288, 1, 1 | F32 | blk.21.ssm_conv1d.weight
+ 404: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_dt.bias
+ 405: 128 | 128, 1, 1, 1 | F32 | blk.21.ssm_norm.weight
+ 406: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.21.ssm_out.weight
+ 407: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.22.attn_gate.weight
+ 408: 3072 | 3072, 1, 1, 1 | F32 | blk.22.attn_norm.weight
+ 409: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.22.attn_qkv.weight
+ 410: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.22.ffn_down_exps.weight
+ 411: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.22.ffn_down_shexp.weight
+ 412: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.22.ffn_gate_exps.weight
+ 413: 786432 | 3072, 256, 1, 1 | F32 | blk.22.ffn_gate_inp.weight
+ 414: 3072 | 3072, 1, 1, 1 | F32 | blk.22.ffn_gate_inp_shexp.weight
+ 415: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_gate_shexp.weight
+ 416: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.22.ffn_up_exps.weight
+ 417: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_up_shexp.weight
+ 418: 3072 | 3072, 1, 1, 1 | F32 | blk.22.post_attention_norm.weight
+ 419: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_a
+ 420: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_alpha.weight
+ 421: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_beta.weight
+ 422: 49152 | 4, 12288, 1, 1 | F32 | blk.22.ssm_conv1d.weight
+ 423: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_dt.bias
+ 424: 128 | 128, 1, 1, 1 | F32 | blk.22.ssm_norm.weight
+ 425: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.22.ssm_out.weight
+ 426: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_k.weight
+ 427: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_k_norm.weight
+ 428: 3072 | 3072, 1, 1, 1 | F32 | blk.23.attn_norm.weight
+ 429: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.23.attn_output.weight
+ 430: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.23.attn_q.weight
+ 431: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_q_norm.weight
+ 432: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_v.weight
+ 433: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.23.ffn_down_exps.weight
+ 434: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.23.ffn_down_shexp.weight
+ 435: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.23.ffn_gate_exps.weight
+ 436: 786432 | 3072, 256, 1, 1 | F32 | blk.23.ffn_gate_inp.weight
+ 437: 3072 | 3072, 1, 1, 1 | F32 | blk.23.ffn_gate_inp_shexp.weight
+ 438: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_gate_shexp.weight
+ 439: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.23.ffn_up_exps.weight
+ 440: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_up_shexp.weight
+ 441: 3072 | 3072, 1, 1, 1 | F32 | blk.23.post_attention_norm.weight
+ 442: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.24.attn_gate.weight
+ 443: 3072 | 3072, 1, 1, 1 | F32 | blk.24.attn_norm.weight
+ 444: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.24.attn_qkv.weight
+ 445: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.24.ffn_down_exps.weight
+ 446: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.24.ffn_down_shexp.weight
+ 447: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.24.ffn_gate_exps.weight
+ 448: 786432 | 3072, 256, 1, 1 | F32 | blk.24.ffn_gate_inp.weight
+ 449: 3072 | 3072, 1, 1, 1 | F32 | blk.24.ffn_gate_inp_shexp.weight
+ 450: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_gate_shexp.weight
+ 451: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.24.ffn_up_exps.weight
+ 452: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_up_shexp.weight
+ 453: 3072 | 3072, 1, 1, 1 | F32 | blk.24.post_attention_norm.weight
+ 454: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_a
+ 455: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_alpha.weight
+ 456: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_beta.weight
+ 457: 49152 | 4, 12288, 1, 1 | F32 | blk.24.ssm_conv1d.weight
+ 458: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_dt.bias
+ 459: 128 | 128, 1, 1, 1 | F32 | blk.24.ssm_norm.weight
+ 460: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.24.ssm_out.weight
+ 461: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.25.attn_gate.weight
+ 462: 3072 | 3072, 1, 1, 1 | F32 | blk.25.attn_norm.weight
+ 463: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.25.attn_qkv.weight
+ 464: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.25.ffn_down_exps.weight
+ 465: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.25.ffn_down_shexp.weight
+ 466: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.25.ffn_gate_exps.weight
+ 467: 786432 | 3072, 256, 1, 1 | F32 | blk.25.ffn_gate_inp.weight
+ 468: 3072 | 3072, 1, 1, 1 | F32 | blk.25.ffn_gate_inp_shexp.weight
+ 469: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_gate_shexp.weight
+ 470: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.25.ffn_up_exps.weight
+ 471: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_up_shexp.weight
+ 472: 3072 | 3072, 1, 1, 1 | F32 | blk.25.post_attention_norm.weight
+ 473: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_a
+ 474: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_alpha.weight
+ 475: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_beta.weight
+ 476: 49152 | 4, 12288, 1, 1 | F32 | blk.25.ssm_conv1d.weight
+ 477: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_dt.bias
+ 478: 128 | 128, 1, 1, 1 | F32 | blk.25.ssm_norm.weight
+ 479: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.25.ssm_out.weight
+ 480: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.26.attn_gate.weight
+ 481: 3072 | 3072, 1, 1, 1 | F32 | blk.26.attn_norm.weight
+ 482: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.26.attn_qkv.weight
+ 483: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.26.ffn_down_exps.weight
+ 484: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.26.ffn_down_shexp.weight
+ 485: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.26.ffn_gate_exps.weight
+ 486: 786432 | 3072, 256, 1, 1 | F32 | blk.26.ffn_gate_inp.weight
+ 487: 3072 | 3072, 1, 1, 1 | F32 | blk.26.ffn_gate_inp_shexp.weight
+ 488: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_gate_shexp.weight
+ 489: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.26.ffn_up_exps.weight
+ 490: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_up_shexp.weight
+ 491: 3072 | 3072, 1, 1, 1 | F32 | blk.26.post_attention_norm.weight
+ 492: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_a
+ 493: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_alpha.weight
+ 494: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_beta.weight
+ 495: 49152 | 4, 12288, 1, 1 | F32 | blk.26.ssm_conv1d.weight
+ 496: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_dt.bias
+ 497: 128 | 128, 1, 1, 1 | F32 | blk.26.ssm_norm.weight
+ 498: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.26.ssm_out.weight
+ 499: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_k.weight
+ 500: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_k_norm.weight
+ 501: 3072 | 3072, 1, 1, 1 | F32 | blk.27.attn_norm.weight
+ 502: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.27.attn_output.weight
+ 503: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.27.attn_q.weight
+ 504: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_q_norm.weight
+ 505: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_v.weight
+ 506: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.27.ffn_down_exps.weight
+ 507: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.27.ffn_down_shexp.weight
+ 508: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.27.ffn_gate_exps.weight
+ 509: 786432 | 3072, 256, 1, 1 | F32 | blk.27.ffn_gate_inp.weight
+ 510: 3072 | 3072, 1, 1, 1 | F32 | blk.27.ffn_gate_inp_shexp.weight
+ 511: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_gate_shexp.weight
+ 512: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.27.ffn_up_exps.weight
+ 513: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_up_shexp.weight
+ 514: 3072 | 3072, 1, 1, 1 | F32 | blk.27.post_attention_norm.weight
+ 515: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.28.attn_gate.weight
+ 516: 3072 | 3072, 1, 1, 1 | F32 | blk.28.attn_norm.weight
+ 517: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.28.attn_qkv.weight
+ 518: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.28.ffn_down_exps.weight
+ 519: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.28.ffn_down_shexp.weight
+ 520: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.28.ffn_gate_exps.weight
+ 521: 786432 | 3072, 256, 1, 1 | F32 | blk.28.ffn_gate_inp.weight
+ 522: 3072 | 3072, 1, 1, 1 | F32 | blk.28.ffn_gate_inp_shexp.weight
+ 523: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_gate_shexp.weight
+ 524: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.28.ffn_up_exps.weight
+ 525: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_up_shexp.weight
+ 526: 3072 | 3072, 1, 1, 1 | F32 | blk.28.post_attention_norm.weight
+ 527: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_a
+ 528: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_alpha.weight
+ 529: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_beta.weight
+ 530: 49152 | 4, 12288, 1, 1 | F32 | blk.28.ssm_conv1d.weight
+ 531: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_dt.bias
+ 532: 128 | 128, 1, 1, 1 | F32 | blk.28.ssm_norm.weight
+ 533: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.28.ssm_out.weight
+ 534: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.29.attn_gate.weight
+ 535: 3072 | 3072, 1, 1, 1 | F32 | blk.29.attn_norm.weight
+ 536: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.29.attn_qkv.weight
+ 537: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.29.ffn_down_exps.weight
+ 538: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.29.ffn_down_shexp.weight
+ 539: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.29.ffn_gate_exps.weight
+ 540: 786432 | 3072, 256, 1, 1 | F32 | blk.29.ffn_gate_inp.weight
+ 541: 3072 | 3072, 1, 1, 1 | F32 | blk.29.ffn_gate_inp_shexp.weight
+ 542: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_gate_shexp.weight
+ 543: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.29.ffn_up_exps.weight
+ 544: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_up_shexp.weight
+ 545: 3072 | 3072, 1, 1, 1 | F32 | blk.29.post_attention_norm.weight
+ 546: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_a
+ 547: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_alpha.weight
+ 548: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_beta.weight
+ 549: 49152 | 4, 12288, 1, 1 | F32 | blk.29.ssm_conv1d.weight
+ 550: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_dt.bias
+ 551: 128 | 128, 1, 1, 1 | F32 | blk.29.ssm_norm.weight
+ 552: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.29.ssm_out.weight
+ 553: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.30.attn_gate.weight
+ 554: 3072 | 3072, 1, 1, 1 | F32 | blk.30.attn_norm.weight
+ 555: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.30.attn_qkv.weight
+ 556: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.30.ffn_down_exps.weight
+ 557: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.30.ffn_down_shexp.weight
+ 558: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.30.ffn_gate_exps.weight
+ 559: 786432 | 3072, 256, 1, 1 | F32 | blk.30.ffn_gate_inp.weight
+ 560: 3072 | 3072, 1, 1, 1 | F32 | blk.30.ffn_gate_inp_shexp.weight
+ 561: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_gate_shexp.weight
+ 562: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.30.ffn_up_exps.weight
+ 563: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_up_shexp.weight
+ 564: 3072 | 3072, 1, 1, 1 | F32 | blk.30.post_attention_norm.weight
+ 565: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_a
+ 566: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_alpha.weight
+ 567: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_beta.weight
+ 568: 49152 | 4, 12288, 1, 1 | F32 | blk.30.ssm_conv1d.weight
+ 569: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_dt.bias
+ 570: 128 | 128, 1, 1, 1 | F32 | blk.30.ssm_norm.weight
+ 571: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.30.ssm_out.weight
+ 572: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_k.weight
+ 573: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_k_norm.weight
+ 574: 3072 | 3072, 1, 1, 1 | F32 | blk.31.attn_norm.weight
+ 575: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.31.attn_output.weight
+ 576: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.31.attn_q.weight
+ 577: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_q_norm.weight
+ 578: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_v.weight
+ 579: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.31.ffn_down_exps.weight
+ 580: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.31.ffn_down_shexp.weight
+ 581: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.31.ffn_gate_exps.weight
+ 582: 786432 | 3072, 256, 1, 1 | F32 | blk.31.ffn_gate_inp.weight
+ 583: 3072 | 3072, 1, 1, 1 | F32 | blk.31.ffn_gate_inp_shexp.weight
+ 584: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_gate_shexp.weight
+ 585: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.31.ffn_up_exps.weight
+ 586: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_up_shexp.weight
+ 587: 3072 | 3072, 1, 1, 1 | F32 | blk.31.post_attention_norm.weight
+ 588: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.32.attn_gate.weight
+ 589: 3072 | 3072, 1, 1, 1 | F32 | blk.32.attn_norm.weight
+ 590: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.32.attn_qkv.weight
+ 591: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.32.ffn_down_exps.weight
+ 592: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.32.ffn_down_shexp.weight
+ 593: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.32.ffn_gate_exps.weight
+ 594: 786432 | 3072, 256, 1, 1 | F32 | blk.32.ffn_gate_inp.weight
+ 595: 3072 | 3072, 1, 1, 1 | F32 | blk.32.ffn_gate_inp_shexp.weight
+ 596: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_gate_shexp.weight
+ 597: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.32.ffn_up_exps.weight
+ 598: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_up_shexp.weight
+ 599: 3072 | 3072, 1, 1, 1 | F32 | blk.32.post_attention_norm.weight
+ 600: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_a
+ 601: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_alpha.weight
+ 602: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_beta.weight
+ 603: 49152 | 4, 12288, 1, 1 | F32 | blk.32.ssm_conv1d.weight
+ 604: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_dt.bias
+ 605: 128 | 128, 1, 1, 1 | F32 | blk.32.ssm_norm.weight
+ 606: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.32.ssm_out.weight
+ 607: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.33.attn_gate.weight
+ 608: 3072 | 3072, 1, 1, 1 | F32 | blk.33.attn_norm.weight
+ 609: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.33.attn_qkv.weight
+ 610: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.33.ffn_down_exps.weight
+ 611: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.33.ffn_down_shexp.weight
+ 612: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.33.ffn_gate_exps.weight
+ 613: 786432 | 3072, 256, 1, 1 | F32 | blk.33.ffn_gate_inp.weight
+ 614: 3072 | 3072, 1, 1, 1 | F32 | blk.33.ffn_gate_inp_shexp.weight
+ 615: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_gate_shexp.weight
+ 616: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.33.ffn_up_exps.weight
+ 617: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_up_shexp.weight
+ 618: 3072 | 3072, 1, 1, 1 | F32 | blk.33.post_attention_norm.weight
+ 619: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_a
+ 620: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_alpha.weight
+ 621: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_beta.weight
+ 622: 49152 | 4, 12288, 1, 1 | F32 | blk.33.ssm_conv1d.weight
+ 623: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_dt.bias
+ 624: 128 | 128, 1, 1, 1 | F32 | blk.33.ssm_norm.weight
+ 625: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.33.ssm_out.weight
+ 626: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.34.attn_gate.weight
+ 627: 3072 | 3072, 1, 1, 1 | F32 | blk.34.attn_norm.weight
+ 628: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.34.attn_qkv.weight
+ 629: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.34.ffn_down_exps.weight
+ 630: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.34.ffn_down_shexp.weight
+ 631: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.34.ffn_gate_exps.weight
+ 632: 786432 | 3072, 256, 1, 1 | F32 | blk.34.ffn_gate_inp.weight
+ 633: 3072 | 3072, 1, 1, 1 | F32 | blk.34.ffn_gate_inp_shexp.weight
+ 634: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_gate_shexp.weight
+ 635: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.34.ffn_up_exps.weight
+ 636: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_up_shexp.weight
+ 637: 3072 | 3072, 1, 1, 1 | F32 | blk.34.post_attention_norm.weight
+ 638: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_a
+ 639: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_alpha.weight
+ 640: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_beta.weight
+ 641: 49152 | 4, 12288, 1, 1 | F32 | blk.34.ssm_conv1d.weight
+ 642: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_dt.bias
+ 643: 128 | 128, 1, 1, 1 | F32 | blk.34.ssm_norm.weight
+ 644: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.34.ssm_out.weight
+ 645: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_k.weight
+ 646: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_k_norm.weight
+ 647: 3072 | 3072, 1, 1, 1 | F32 | blk.35.attn_norm.weight
+ 648: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.35.attn_output.weight
+ 649: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.35.attn_q.weight
+ 650: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_q_norm.weight
+ 651: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_v.weight
+ 652: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.35.ffn_down_exps.weight
+ 653: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.35.ffn_down_shexp.weight
+ 654: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.35.ffn_gate_exps.weight
+ 655: 786432 | 3072, 256, 1, 1 | F32 | blk.35.ffn_gate_inp.weight
+ 656: 3072 | 3072, 1, 1, 1 | F32 | blk.35.ffn_gate_inp_shexp.weight
+ 657: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_gate_shexp.weight
+ 658: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.35.ffn_up_exps.weight
+ 659: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_up_shexp.weight
+ 660: 3072 | 3072, 1, 1, 1 | F32 | blk.35.post_attention_norm.weight
+ 661: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.36.attn_gate.weight
+ 662: 3072 | 3072, 1, 1, 1 | F32 | blk.36.attn_norm.weight
+ 663: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.36.attn_qkv.weight
+ 664: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.36.ffn_down_exps.weight
+ 665: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.36.ffn_down_shexp.weight
+ 666: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.36.ffn_gate_exps.weight
+ 667: 786432 | 3072, 256, 1, 1 | F32 | blk.36.ffn_gate_inp.weight
+ 668: 3072 | 3072, 1, 1, 1 | F32 | blk.36.ffn_gate_inp_shexp.weight
+ 669: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_gate_shexp.weight
+ 670: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.36.ffn_up_exps.weight
+ 671: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_up_shexp.weight
+ 672: 3072 | 3072, 1, 1, 1 | F32 | blk.36.post_attention_norm.weight
+ 673: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_a
+ 674: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_alpha.weight
+ 675: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_beta.weight
+ 676: 49152 | 4, 12288, 1, 1 | F32 | blk.36.ssm_conv1d.weight
+ 677: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_dt.bias
+ 678: 128 | 128, 1, 1, 1 | F32 | blk.36.ssm_norm.weight
+ 679: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.36.ssm_out.weight
+ 680: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.37.attn_gate.weight
+ 681: 3072 | 3072, 1, 1, 1 | F32 | blk.37.attn_norm.weight
+ 682: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.37.attn_qkv.weight
+ 683: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.37.ffn_down_exps.weight
+ 684: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.37.ffn_down_shexp.weight
+ 685: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.37.ffn_gate_exps.weight
+ 686: 786432 | 3072, 256, 1, 1 | F32 | blk.37.ffn_gate_inp.weight
+ 687: 3072 | 3072, 1, 1, 1 | F32 | blk.37.ffn_gate_inp_shexp.weight
+ 688: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_gate_shexp.weight
+ 689: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.37.ffn_up_exps.weight
+ 690: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_up_shexp.weight
+ 691: 3072 | 3072, 1, 1, 1 | F32 | blk.37.post_attention_norm.weight
+ 692: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_a
+ 693: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_alpha.weight
+ 694: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_beta.weight
+ 695: 49152 | 4, 12288, 1, 1 | F32 | blk.37.ssm_conv1d.weight
+ 696: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_dt.bias
+ 697: 128 | 128, 1, 1, 1 | F32 | blk.37.ssm_norm.weight
+ 698: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.37.ssm_out.weight
+ 699: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.38.attn_gate.weight
+ 700: 3072 | 3072, 1, 1, 1 | F32 | blk.38.attn_norm.weight
+ 701: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.38.attn_qkv.weight
+ 702: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.38.ffn_down_exps.weight
+ 703: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.38.ffn_down_shexp.weight
+ 704: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.38.ffn_gate_exps.weight
+ 705: 786432 | 3072, 256, 1, 1 | F32 | blk.38.ffn_gate_inp.weight
+ 706: 3072 | 3072, 1, 1, 1 | F32 | blk.38.ffn_gate_inp_shexp.weight
+ 707: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_gate_shexp.weight
+ 708: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.38.ffn_up_exps.weight
+ 709: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_up_shexp.weight
+ 710: 3072 | 3072, 1, 1, 1 | F32 | blk.38.post_attention_norm.weight
+ 711: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_a
+ 712: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_alpha.weight
+ 713: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_beta.weight
+ 714: 49152 | 4, 12288, 1, 1 | F32 | blk.38.ssm_conv1d.weight
+ 715: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_dt.bias
+ 716: 128 | 128, 1, 1, 1 | F32 | blk.38.ssm_norm.weight
+ 717: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.38.ssm_out.weight
+ 718: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_k.weight
+ 719: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_k_norm.weight
+ 720: 3072 | 3072, 1, 1, 1 | F32 | blk.39.attn_norm.weight
+ 721: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.39.attn_output.weight
+ 722: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.39.attn_q.weight
+ 723: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_q_norm.weight
+ 724: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_v.weight
+ 725: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.39.ffn_down_exps.weight
+ 726: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.39.ffn_down_shexp.weight
+ 727: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.39.ffn_gate_exps.weight
+ 728: 786432 | 3072, 256, 1, 1 | F32 | blk.39.ffn_gate_inp.weight
+ 729: 3072 | 3072, 1, 1, 1 | F32 | blk.39.ffn_gate_inp_shexp.weight
+ 730: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_gate_shexp.weight
+ 731: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.39.ffn_up_exps.weight
+ 732: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_up_shexp.weight
+ 733: 3072 | 3072, 1, 1, 1 | F32 | blk.39.post_attention_norm.weight
+ 734: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.40.attn_gate.weight
+ 735: 3072 | 3072, 1, 1, 1 | F32 | blk.40.attn_norm.weight
+ 736: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.40.attn_qkv.weight
+ 737: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.40.ffn_down_exps.weight
+ 738: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.40.ffn_down_shexp.weight
+ 739: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.40.ffn_gate_exps.weight
+ 740: 786432 | 3072, 256, 1, 1 | F32 | blk.40.ffn_gate_inp.weight
+ 741: 3072 | 3072, 1, 1, 1 | F32 | blk.40.ffn_gate_inp_shexp.weight
+ 742: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_gate_shexp.weight
+ 743: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.40.ffn_up_exps.weight
+ 744: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_up_shexp.weight
+ 745: 3072 | 3072, 1, 1, 1 | F32 | blk.40.post_attention_norm.weight
+ 746: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_a
+ 747: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_alpha.weight
+ 748: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_beta.weight
+ 749: 49152 | 4, 12288, 1, 1 | F32 | blk.40.ssm_conv1d.weight
+ 750: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_dt.bias
+ 751: 128 | 128, 1, 1, 1 | F32 | blk.40.ssm_norm.weight
+ 752: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.40.ssm_out.weight
+ 753: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.41.attn_gate.weight
+ 754: 3072 | 3072, 1, 1, 1 | F32 | blk.41.attn_norm.weight
+ 755: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.41.attn_qkv.weight
+ 756: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.41.ffn_down_exps.weight
+ 757: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.41.ffn_down_shexp.weight
+ 758: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.41.ffn_gate_exps.weight
+ 759: 786432 | 3072, 256, 1, 1 | F32 | blk.41.ffn_gate_inp.weight
+ 760: 3072 | 3072, 1, 1, 1 | F32 | blk.41.ffn_gate_inp_shexp.weight
+ 761: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_gate_shexp.weight
+ 762: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.41.ffn_up_exps.weight
+ 763: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_up_shexp.weight
+ 764: 3072 | 3072, 1, 1, 1 | F32 | blk.41.post_attention_norm.weight
+ 765: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_a
+ 766: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_alpha.weight
+ 767: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_beta.weight
+ 768: 49152 | 4, 12288, 1, 1 | F32 | blk.41.ssm_conv1d.weight
+ 769: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_dt.bias
+ 770: 128 | 128, 1, 1, 1 | F32 | blk.41.ssm_norm.weight
+ 771: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.41.ssm_out.weight
+ 772: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.42.attn_gate.weight
+ 773: 3072 | 3072, 1, 1, 1 | F32 | blk.42.attn_norm.weight
+ 774: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.42.attn_qkv.weight
+ 775: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.42.ffn_down_exps.weight
+ 776: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.42.ffn_down_shexp.weight
+ 777: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.42.ffn_gate_exps.weight
+ 778: 786432 | 3072, 256, 1, 1 | F32 | blk.42.ffn_gate_inp.weight
+ 779: 3072 | 3072, 1, 1, 1 | F32 | blk.42.ffn_gate_inp_shexp.weight
+ 780: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_gate_shexp.weight
+ 781: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.42.ffn_up_exps.weight
+ 782: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_up_shexp.weight
+ 783: 3072 | 3072, 1, 1, 1 | F32 | blk.42.post_attention_norm.weight
+ 784: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_a
+ 785: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_alpha.weight
+ 786: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_beta.weight
+ 787: 49152 | 4, 12288, 1, 1 | F32 | blk.42.ssm_conv1d.weight
+ 788: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_dt.bias
+ 789: 128 | 128, 1, 1, 1 | F32 | blk.42.ssm_norm.weight
+ 790: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.42.ssm_out.weight
+ 791: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_k.weight
+ 792: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_k_norm.weight
+ 793: 3072 | 3072, 1, 1, 1 | F32 | blk.43.attn_norm.weight
+ 794: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.43.attn_output.weight
+ 795: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.43.attn_q.weight
+ 796: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_q_norm.weight
+ 797: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_v.weight
+ 798: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.43.ffn_down_exps.weight
+ 799: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.43.ffn_down_shexp.weight
+ 800: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.43.ffn_gate_exps.weight
+ 801: 786432 | 3072, 256, 1, 1 | F32 | blk.43.ffn_gate_inp.weight
+ 802: 3072 | 3072, 1, 1, 1 | F32 | blk.43.ffn_gate_inp_shexp.weight
+ 803: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_gate_shexp.weight
+ 804: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.43.ffn_up_exps.weight
+ 805: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_up_shexp.weight
+ 806: 3072 | 3072, 1, 1, 1 | F32 | blk.43.post_attention_norm.weight
+ 807: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.44.attn_gate.weight
+ 808: 3072 | 3072, 1, 1, 1 | F32 | blk.44.attn_norm.weight
+ 809: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.44.attn_qkv.weight
+ 810: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.44.ffn_down_exps.weight
+ 811: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.44.ffn_down_shexp.weight
+ 812: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.44.ffn_gate_exps.weight
+ 813: 786432 | 3072, 256, 1, 1 | F32 | blk.44.ffn_gate_inp.weight
+ 814: 3072 | 3072, 1, 1, 1 | F32 | blk.44.ffn_gate_inp_shexp.weight
+ 815: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_gate_shexp.weight
+ 816: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.44.ffn_up_exps.weight
+ 817: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_up_shexp.weight
+ 818: 3072 | 3072, 1, 1, 1 | F32 | blk.44.post_attention_norm.weight
+ 819: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_a
+ 820: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_alpha.weight
+ 821: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_beta.weight
+ 822: 49152 | 4, 12288, 1, 1 | F32 | blk.44.ssm_conv1d.weight
+ 823: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_dt.bias
+ 824: 128 | 128, 1, 1, 1 | F32 | blk.44.ssm_norm.weight
+ 825: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.44.ssm_out.weight
+ 826: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.45.attn_gate.weight
+ 827: 3072 | 3072, 1, 1, 1 | F32 | blk.45.attn_norm.weight
+ 828: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.45.attn_qkv.weight
+ 829: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.45.ffn_down_exps.weight
+ 830: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.45.ffn_down_shexp.weight
+ 831: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.45.ffn_gate_exps.weight
+ 832: 786432 | 3072, 256, 1, 1 | F32 | blk.45.ffn_gate_inp.weight
+ 833: 3072 | 3072, 1, 1, 1 | F32 | blk.45.ffn_gate_inp_shexp.weight
+ 834: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_gate_shexp.weight
+ 835: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.45.ffn_up_exps.weight
+ 836: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_up_shexp.weight
+ 837: 3072 | 3072, 1, 1, 1 | F32 | blk.45.post_attention_norm.weight
+ 838: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_a
+ 839: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_alpha.weight
+ 840: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_beta.weight
+ 841: 49152 | 4, 12288, 1, 1 | F32 | blk.45.ssm_conv1d.weight
+ 842: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_dt.bias
+ 843: 128 | 128, 1, 1, 1 | F32 | blk.45.ssm_norm.weight
+ 844: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.45.ssm_out.weight
+ 845: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.46.attn_gate.weight
+ 846: 3072 | 3072, 1, 1, 1 | F32 | blk.46.attn_norm.weight
+ 847: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.46.attn_qkv.weight
+ 848: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.46.ffn_down_exps.weight
+ 849: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.46.ffn_down_shexp.weight
+ 850: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.46.ffn_gate_exps.weight
+ 851: 786432 | 3072, 256, 1, 1 | F32 | blk.46.ffn_gate_inp.weight
+ 852: 3072 | 3072, 1, 1, 1 | F32 | blk.46.ffn_gate_inp_shexp.weight
+ 853: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_gate_shexp.weight
+ 854: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.46.ffn_up_exps.weight
+ 855: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_up_shexp.weight
+ 856: 3072 | 3072, 1, 1, 1 | F32 | blk.46.post_attention_norm.weight
+ 857: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_a
+ 858: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_alpha.weight
+ 859: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_beta.weight
+ 860: 49152 | 4, 12288, 1, 1 | F32 | blk.46.ssm_conv1d.weight
+ 861: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_dt.bias
+ 862: 128 | 128, 1, 1, 1 | F32 | blk.46.ssm_norm.weight
+ 863: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.46.ssm_out.weight
+ 864: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_k.weight
+ 865: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_k_norm.weight
+ 866: 3072 | 3072, 1, 1, 1 | F32 | blk.47.attn_norm.weight
+ 867: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.47.attn_output.weight
+ 868: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.47.attn_q.weight
+ 869: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_q_norm.weight
+ 870: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_v.weight
+ 871: 805306368 | 1024, 3072, 256, 1 | Q6_K | blk.47.ffn_down_exps.weight
+ 872: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.47.ffn_down_shexp.weight
+ 873: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.47.ffn_gate_exps.weight
+ 874: 786432 | 3072, 256, 1, 1 | F32 | blk.47.ffn_gate_inp.weight
+ 875: 3072 | 3072, 1, 1, 1 | F32 | blk.47.ffn_gate_inp_shexp.weight
+ 876: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_gate_shexp.weight
+ 877: 805306368 | 3072, 1024, 256, 1 | Q6_K | blk.47.ffn_up_exps.weight
+ 878: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_up_shexp.weight
+ 879: 3072 | 3072, 1, 1, 1 | F32 | blk.47.post_attention_norm.weight
+
+
\ No newline at end of file