Iceberg-Yang commited on
Commit
e63c55a
·
verified ·
1 Parent(s): 9dd5bd2

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. chat_template.jinja +159 -0
  2. config.json +1109 -0
  3. generation_config.json +12 -0
  4. merges.txt +0 -0
  5. model-00001-of-00131.safetensors +3 -0
  6. model-00002-of-00131.safetensors +3 -0
  7. model-00003-of-00131.safetensors +3 -0
  8. model-00004-of-00131.safetensors +3 -0
  9. model-00005-of-00131.safetensors +3 -0
  10. model-00006-of-00131.safetensors +3 -0
  11. model-00007-of-00131.safetensors +3 -0
  12. model-00008-of-00131.safetensors +3 -0
  13. model-00009-of-00131.safetensors +3 -0
  14. model-00010-of-00131.safetensors +3 -0
  15. model-00011-of-00131.safetensors +3 -0
  16. model-00012-of-00131.safetensors +3 -0
  17. model-00013-of-00131.safetensors +3 -0
  18. model-00014-of-00131.safetensors +3 -0
  19. model-00015-of-00131.safetensors +3 -0
  20. model-00016-of-00131.safetensors +3 -0
  21. model-00017-of-00131.safetensors +3 -0
  22. model-00018-of-00131.safetensors +3 -0
  23. model-00019-of-00131.safetensors +3 -0
  24. model-00020-of-00131.safetensors +3 -0
  25. model-00021-of-00131.safetensors +3 -0
  26. model-00022-of-00131.safetensors +3 -0
  27. model-00023-of-00131.safetensors +3 -0
  28. model-00024-of-00131.safetensors +3 -0
  29. model-00025-of-00131.safetensors +3 -0
  30. model-00026-of-00131.safetensors +3 -0
  31. model-00027-of-00131.safetensors +3 -0
  32. model-00028-of-00131.safetensors +3 -0
  33. model-00029-of-00131.safetensors +3 -0
  34. model-00030-of-00131.safetensors +3 -0
  35. model-00031-of-00131.safetensors +3 -0
  36. model-00032-of-00131.safetensors +3 -0
  37. model-00033-of-00131.safetensors +3 -0
  38. model-00034-of-00131.safetensors +3 -0
  39. model-00035-of-00131.safetensors +3 -0
  40. model-00036-of-00131.safetensors +3 -0
  41. model-00037-of-00131.safetensors +3 -0
  42. model-00038-of-00131.safetensors +3 -0
  43. model-00039-of-00131.safetensors +3 -0
  44. model-00040-of-00131.safetensors +3 -0
  45. model-00041-of-00131.safetensors +3 -0
  46. model-00042-of-00131.safetensors +3 -0
  47. model-00043-of-00131.safetensors +3 -0
  48. model-00044-of-00131.safetensors +3 -0
  49. model-00045-of-00131.safetensors +3 -0
  50. model-00046-of-00131.safetensors +3 -0
chat_template.jinja ADDED
@@ -0,0 +1,159 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- set image_count = namespace(value=0) %}
2
+ {%- set video_count = namespace(value=0) %}
3
+ {%- macro render_content(content, do_vision_count, is_system_content=false) %}
4
+ {%- if content is string %}
5
+ {{- content }}
6
+ {%- elif content is iterable and content is not mapping %}
7
+ {%- for item in content %}
8
+ {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
9
+ {%- if is_system_content %}
10
+ {{- raise_exception('System message cannot contain images.') }}
11
+ {%- endif %}
12
+ {%- if do_vision_count %}
13
+ {%- set image_count.value = image_count.value + 1 %}
14
+ {%- endif %}
15
+ {%- if add_vision_id %}
16
+ {{- 'Picture ' ~ image_count.value ~ ': ' }}
17
+ {%- endif %}
18
+ {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
19
+ {%- elif 'video' in item or item.type == 'video' %}
20
+ {%- if is_system_content %}
21
+ {{- raise_exception('System message cannot contain videos.') }}
22
+ {%- endif %}
23
+ {%- if do_vision_count %}
24
+ {%- set video_count.value = video_count.value + 1 %}
25
+ {%- endif %}
26
+ {%- if add_vision_id %}
27
+ {{- 'Video ' ~ video_count.value ~ ': ' }}
28
+ {%- endif %}
29
+ {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
30
+ {%- elif 'audio' in item or item.type == 'audio' %}
31
+ {%- if is_system_content %}
32
+ {{- raise_exception('System message cannot contain audio.') }}
33
+ {%- endif %}
34
+ {{- '<|audio_start|><|audio_pad|><|audio_end|>' }}
35
+ {%- elif 'text' in item %}
36
+ {{- item.text }}
37
+ {%- else %}
38
+ {{- raise_exception('Unexpected item type in content.') }}
39
+ {%- endif %}
40
+ {%- endfor %}
41
+ {%- elif content is none or content is undefined %}
42
+ {{- '' }}
43
+ {%- else %}
44
+ {{- raise_exception('Unexpected content type.') }}
45
+ {%- endif %}
46
+ {%- endmacro %}
47
+ {%- if not messages %}
48
+ {{- raise_exception('No messages provided.') }}
49
+ {%- endif %}
50
+ {%- if tools and tools is iterable and tools is not mapping %}
51
+ {{- '<|im_start|>system\n' }}
52
+ {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
53
+ {%- for tool in tools %}
54
+ {{- "\n" }}
55
+ {{- tool | tojson }}
56
+ {%- endfor %}
57
+ {{- "\n</tools>" }}
58
+ {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
59
+ {%- if messages[0].role == 'system' %}
60
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
61
+ {%- if content %}
62
+ {{- '\n\n' + content }}
63
+ {%- endif %}
64
+ {%- endif %}
65
+ {{- '<|im_end|>\n' }}
66
+ {%- else %}
67
+ {%- if messages[0].role == 'system' %}
68
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
69
+ {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
70
+ {%- endif %}
71
+ {%- endif %}
72
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
73
+ {%- for message in messages[::-1] %}
74
+ {%- set index = (messages|length - 1) - loop.index0 %}
75
+ {%- if ns.multi_step_tool and message.role == "user" %}
76
+ {%- set content = render_content(message.content, false)|trim %}
77
+ {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
78
+ {%- set ns.multi_step_tool = false %}
79
+ {%- set ns.last_query_index = index %}
80
+ {%- endif %}
81
+ {%- endif %}
82
+ {%- endfor %}
83
+ {%- if ns.multi_step_tool %}
84
+ {{- raise_exception('No user query found in messages.') }}
85
+ {%- endif %}
86
+ {%- for message in messages %}
87
+ {%- set content = render_content(message.content, true)|trim %}
88
+ {%- if message.role == "system" %}
89
+ {%- if not loop.first %}
90
+ {{- raise_exception('System message must be at the beginning.') }}
91
+ {%- endif %}
92
+ {%- elif message.role == "user" %}
93
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
94
+ {%- elif message.role == "assistant" %}
95
+ {%- set reasoning_content = '' %}
96
+ {%- if message.reasoning_content is string %}
97
+ {%- set reasoning_content = message.reasoning_content %}
98
+ {%- else %}
99
+ {%- if '</think>' in content %}
100
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
101
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
102
+ {%- endif %}
103
+ {%- endif %}
104
+ {%- set reasoning_content = reasoning_content|trim %}
105
+ {%- if (preserve_thinking is defined and preserve_thinking is true) or (loop.index0 > ns.last_query_index) %}
106
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
107
+ {%- else %}
108
+ {{- '<|im_start|>' + message.role + '\n' + content }}
109
+ {%- endif %}
110
+ {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
111
+ {%- for tool_call in message.tool_calls %}
112
+ {%- if tool_call.function is defined %}
113
+ {%- set tool_call = tool_call.function %}
114
+ {%- endif %}
115
+ {%- if loop.first %}
116
+ {%- if content|trim %}
117
+ {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
118
+ {%- else %}
119
+ {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
120
+ {%- endif %}
121
+ {%- else %}
122
+ {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
123
+ {%- endif %}
124
+ {%- if tool_call.arguments is defined %}
125
+ {%- for args_name, args_value in tool_call.arguments|items %}
126
+ {{- '<parameter=' + args_name + '>\n' }}
127
+ {%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}
128
+ {{- args_value }}
129
+ {{- '\n</parameter>\n' }}
130
+ {%- endfor %}
131
+ {%- endif %}
132
+ {{- '</function>\n</tool_call>' }}
133
+ {%- endfor %}
134
+ {%- endif %}
135
+ {{- '<|im_end|>\n' }}
136
+ {%- elif message.role == "tool" %}
137
+ {%- if loop.previtem and loop.previtem.role != "tool" %}
138
+ {{- '<|im_start|>user' }}
139
+ {%- endif %}
140
+ {{- '\n<tool_response>\n' }}
141
+ {{- content }}
142
+ {{- '\n</tool_response>' }}
143
+ {%- if not loop.last and loop.nextitem.role != "tool" %}
144
+ {{- '<|im_end|>\n' }}
145
+ {%- elif loop.last %}
146
+ {{- '<|im_end|>\n' }}
147
+ {%- endif %}
148
+ {%- else %}
149
+ {{- raise_exception('Unexpected message role.') }}
150
+ {%- endif %}
151
+ {%- endfor %}
152
+ {%- if add_generation_prompt %}
153
+ {{- '<|im_start|>assistant\n' }}
154
+ {%- if enable_thinking is defined and enable_thinking is false %}
155
+ {{- '<think>\n\n</think>\n\n' }}
156
+ {%- else %}
157
+ {{- '<think>\n' }}
158
+ {%- endif %}
159
+ {%- endif %}
config.json ADDED
@@ -0,0 +1,1109 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen4ExpForConditionalGeneration"
4
+ ],
5
+ "image_token_id": 248056,
6
+ "language_model_only": false,
7
+ "model_type": "qwen4_exp",
8
+ "text_config": {
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "bos_token_id": 248044,
12
+ "dtype": "bfloat16",
13
+ "eos_token_id": 248044,
14
+ "full_attention_interval": 4,
15
+ "hc_count": 4,
16
+ "hc_lowrank": 320,
17
+ "head_dim": 256,
18
+ "heads_per_ngram": 8,
19
+ "hidden_act": "silu",
20
+ "hidden_size": 2560,
21
+ "indexer_budget": 2048,
22
+ "indexer_compress_ratio": 4,
23
+ "indexer_head_dim": 128,
24
+ "indexer_kv_heads": 1,
25
+ "indexer_n_heads": 4,
26
+ "initializer_range": 0.02,
27
+ "layer_types": [
28
+ "linear_attention",
29
+ "linear_attention",
30
+ "linear_attention",
31
+ "full_attention",
32
+ "linear_attention",
33
+ "linear_attention",
34
+ "linear_attention",
35
+ "full_attention",
36
+ "linear_attention",
37
+ "linear_attention",
38
+ "linear_attention",
39
+ "full_attention",
40
+ "linear_attention",
41
+ "linear_attention",
42
+ "linear_attention",
43
+ "full_attention",
44
+ "linear_attention",
45
+ "linear_attention",
46
+ "linear_attention",
47
+ "full_attention",
48
+ "linear_attention",
49
+ "linear_attention",
50
+ "linear_attention",
51
+ "full_attention",
52
+ "linear_attention",
53
+ "linear_attention",
54
+ "linear_attention",
55
+ "full_attention",
56
+ "linear_attention",
57
+ "linear_attention",
58
+ "linear_attention",
59
+ "full_attention",
60
+ "linear_attention",
61
+ "linear_attention",
62
+ "linear_attention",
63
+ "full_attention",
64
+ "linear_attention",
65
+ "linear_attention",
66
+ "linear_attention",
67
+ "full_attention",
68
+ "linear_attention",
69
+ "linear_attention",
70
+ "linear_attention",
71
+ "full_attention",
72
+ "linear_attention",
73
+ "linear_attention",
74
+ "linear_attention",
75
+ "full_attention"
76
+ ],
77
+ "linear_conv_kernel_dim": 4,
78
+ "linear_key_head_dim": 128,
79
+ "linear_num_key_heads": 16,
80
+ "linear_num_value_heads": 48,
81
+ "linear_value_head_dim": 128,
82
+ "make_ngram_vocab_size_divisible_by": 128,
83
+ "mamba_ssm_dtype": "float32",
84
+ "max_position_embeddings": 262144,
85
+ "model_type": "qwen4_exp_text",
86
+ "moe_intermediate_size": 640,
87
+ "mtp": {
88
+ "hybrid": true,
89
+ "layer_types": [
90
+ "full_attention"
91
+ ],
92
+ "mtp_use_hidden_state_from_layer": null,
93
+ "num_hidden_layers": 1,
94
+ "rope_theta": 10000000
95
+ },
96
+ "mtp_num_hidden_layers": 1,
97
+ "mtp_use_dedicated_embeddings": false,
98
+ "ngram_size": 3,
99
+ "ngram_vocab_size_base": 20000000,
100
+ "num_attention_heads": 24,
101
+ "num_experts": 512,
102
+ "num_experts_per_tok": 10,
103
+ "num_hidden_layers": 48,
104
+ "num_key_value_heads": 2,
105
+ "output_gate_type": "sigmoid",
106
+ "output_router_logits": false,
107
+ "pad_token_id": null,
108
+ "partial_rotary_factor": 0.25,
109
+ "ple_conv_kernel_size": 4,
110
+ "ple_embed_dim": 2560,
111
+ "ple_layer_ids": [
112
+ 2
113
+ ],
114
+ "rms_norm_eps": 1e-06,
115
+ "rope_parameters": {
116
+ "mrope_interleaved": true,
117
+ "mrope_section": [
118
+ 11,
119
+ 11,
120
+ 10
121
+ ],
122
+ "partial_rotary_factor": 0.25,
123
+ "rope_theta": 10000000,
124
+ "rope_type": "default"
125
+ },
126
+ "router_aux_loss_coef": 0.001,
127
+ "shared_expert_intermediate_size": 640,
128
+ "split_ngram_parts": 128,
129
+ "tie_word_embeddings": false,
130
+ "use_cache": true,
131
+ "vocab_size": 248320
132
+ },
133
+ "tie_word_embeddings": false,
134
+ "transformers_version": "5.8.0.dev0",
135
+ "video_token_id": 248057,
136
+ "vision_config": {
137
+ "deepstack_visual_indexes": [],
138
+ "depth": 27,
139
+ "hidden_act": "gelu_pytorch_tanh",
140
+ "hidden_size": 1152,
141
+ "in_channels": 3,
142
+ "initializer_range": 0.02,
143
+ "intermediate_size": 4304,
144
+ "model_type": "qwen4_exp",
145
+ "num_heads": 16,
146
+ "num_position_embeddings": 2304,
147
+ "out_hidden_size": 2560,
148
+ "patch_size": 16,
149
+ "spatial_merge_size": 2,
150
+ "temporal_patch_size": 2
151
+ },
152
+ "vision_end_token_id": 248054,
153
+ "vision_start_token_id": 248053,
154
+ "quantization_config": {
155
+ "quant_method": "fp8",
156
+ "activation_scheme": "dynamic",
157
+ "weight_per_tensor": false,
158
+ "act_per_tensor": false,
159
+ "weight_block_size": [
160
+ 128,
161
+ 128
162
+ ],
163
+ "modules_to_not_convert": [
164
+ "lm_head",
165
+ "model.language_model.embed_tokens",
166
+ "model.language_model.hyper_connection_mixer.input_mix_weight_down",
167
+ "model.language_model.hyper_connection_mixer.input_mix_weight_up",
168
+ "model.language_model.layers.0.attn_hyper_connection.block_inject_weight",
169
+ "model.language_model.layers.0.attn_hyper_connection.input_mix_weight_down",
170
+ "model.language_model.layers.0.attn_hyper_connection.input_mix_weight_up",
171
+ "model.language_model.layers.0.linear_attn.conv1d",
172
+ "model.language_model.layers.0.linear_attn.in_proj_a",
173
+ "model.language_model.layers.0.linear_attn.in_proj_b",
174
+ "model.language_model.layers.0.linear_attn.in_proj_qkv",
175
+ "model.language_model.layers.0.linear_attn.in_proj_z",
176
+ "model.language_model.layers.0.linear_attn.out_proj",
177
+ "model.language_model.layers.0.mlp.gate",
178
+ "model.language_model.layers.0.mlp.shared_expert.down_proj",
179
+ "model.language_model.layers.0.mlp.shared_expert.gate_proj",
180
+ "model.language_model.layers.0.mlp.shared_expert.up_proj",
181
+ "model.language_model.layers.0.mlp.shared_expert_gate",
182
+ "model.language_model.layers.0.mlp_hyper_connection.block_inject_weight",
183
+ "model.language_model.layers.0.mlp_hyper_connection.input_mix_weight_down",
184
+ "model.language_model.layers.0.mlp_hyper_connection.input_mix_weight_up",
185
+ "model.language_model.layers.1.attn_hyper_connection.block_inject_weight",
186
+ "model.language_model.layers.1.attn_hyper_connection.input_mix_weight_down",
187
+ "model.language_model.layers.1.attn_hyper_connection.input_mix_weight_up",
188
+ "model.language_model.layers.1.linear_attn.conv1d",
189
+ "model.language_model.layers.1.linear_attn.in_proj_a",
190
+ "model.language_model.layers.1.linear_attn.in_proj_b",
191
+ "model.language_model.layers.1.linear_attn.in_proj_qkv",
192
+ "model.language_model.layers.1.linear_attn.in_proj_z",
193
+ "model.language_model.layers.1.linear_attn.out_proj",
194
+ "model.language_model.layers.1.mlp.gate",
195
+ "model.language_model.layers.1.mlp.shared_expert.down_proj",
196
+ "model.language_model.layers.1.mlp.shared_expert.gate_proj",
197
+ "model.language_model.layers.1.mlp.shared_expert.up_proj",
198
+ "model.language_model.layers.1.mlp.shared_expert_gate",
199
+ "model.language_model.layers.1.mlp_hyper_connection.block_inject_weight",
200
+ "model.language_model.layers.1.mlp_hyper_connection.input_mix_weight_down",
201
+ "model.language_model.layers.1.mlp_hyper_connection.input_mix_weight_up",
202
+ "model.language_model.layers.1.ple.conv1d",
203
+ "model.language_model.layers.1.ple.key_proj",
204
+ "model.language_model.layers.1.ple.value_proj",
205
+ "model.language_model.layers.10.attn_hyper_connection.block_inject_weight",
206
+ "model.language_model.layers.10.attn_hyper_connection.input_mix_weight_down",
207
+ "model.language_model.layers.10.attn_hyper_connection.input_mix_weight_up",
208
+ "model.language_model.layers.10.linear_attn.conv1d",
209
+ "model.language_model.layers.10.linear_attn.in_proj_a",
210
+ "model.language_model.layers.10.linear_attn.in_proj_b",
211
+ "model.language_model.layers.10.linear_attn.in_proj_qkv",
212
+ "model.language_model.layers.10.linear_attn.in_proj_z",
213
+ "model.language_model.layers.10.linear_attn.out_proj",
214
+ "model.language_model.layers.10.mlp.gate",
215
+ "model.language_model.layers.10.mlp.shared_expert.down_proj",
216
+ "model.language_model.layers.10.mlp.shared_expert.gate_proj",
217
+ "model.language_model.layers.10.mlp.shared_expert.up_proj",
218
+ "model.language_model.layers.10.mlp.shared_expert_gate",
219
+ "model.language_model.layers.10.mlp_hyper_connection.block_inject_weight",
220
+ "model.language_model.layers.10.mlp_hyper_connection.input_mix_weight_down",
221
+ "model.language_model.layers.10.mlp_hyper_connection.input_mix_weight_up",
222
+ "model.language_model.layers.11.attn_hyper_connection.block_inject_weight",
223
+ "model.language_model.layers.11.attn_hyper_connection.input_mix_weight_down",
224
+ "model.language_model.layers.11.attn_hyper_connection.input_mix_weight_up",
225
+ "model.language_model.layers.11.mlp.gate",
226
+ "model.language_model.layers.11.mlp.shared_expert.down_proj",
227
+ "model.language_model.layers.11.mlp.shared_expert.gate_proj",
228
+ "model.language_model.layers.11.mlp.shared_expert.up_proj",
229
+ "model.language_model.layers.11.mlp.shared_expert_gate",
230
+ "model.language_model.layers.11.mlp_hyper_connection.block_inject_weight",
231
+ "model.language_model.layers.11.mlp_hyper_connection.input_mix_weight_down",
232
+ "model.language_model.layers.11.mlp_hyper_connection.input_mix_weight_up",
233
+ "model.language_model.layers.11.self_attn.indexer.index_qk_proj",
234
+ "model.language_model.layers.11.self_attn.k_proj",
235
+ "model.language_model.layers.11.self_attn.o_proj",
236
+ "model.language_model.layers.11.self_attn.q_proj",
237
+ "model.language_model.layers.11.self_attn.v_proj",
238
+ "model.language_model.layers.12.attn_hyper_connection.block_inject_weight",
239
+ "model.language_model.layers.12.attn_hyper_connection.input_mix_weight_down",
240
+ "model.language_model.layers.12.attn_hyper_connection.input_mix_weight_up",
241
+ "model.language_model.layers.12.linear_attn.conv1d",
242
+ "model.language_model.layers.12.linear_attn.in_proj_a",
243
+ "model.language_model.layers.12.linear_attn.in_proj_b",
244
+ "model.language_model.layers.12.linear_attn.in_proj_qkv",
245
+ "model.language_model.layers.12.linear_attn.in_proj_z",
246
+ "model.language_model.layers.12.linear_attn.out_proj",
247
+ "model.language_model.layers.12.mlp.gate",
248
+ "model.language_model.layers.12.mlp.shared_expert.down_proj",
249
+ "model.language_model.layers.12.mlp.shared_expert.gate_proj",
250
+ "model.language_model.layers.12.mlp.shared_expert.up_proj",
251
+ "model.language_model.layers.12.mlp.shared_expert_gate",
252
+ "model.language_model.layers.12.mlp_hyper_connection.block_inject_weight",
253
+ "model.language_model.layers.12.mlp_hyper_connection.input_mix_weight_down",
254
+ "model.language_model.layers.12.mlp_hyper_connection.input_mix_weight_up",
255
+ "model.language_model.layers.13.attn_hyper_connection.block_inject_weight",
256
+ "model.language_model.layers.13.attn_hyper_connection.input_mix_weight_down",
257
+ "model.language_model.layers.13.attn_hyper_connection.input_mix_weight_up",
258
+ "model.language_model.layers.13.linear_attn.conv1d",
259
+ "model.language_model.layers.13.linear_attn.in_proj_a",
260
+ "model.language_model.layers.13.linear_attn.in_proj_b",
261
+ "model.language_model.layers.13.linear_attn.in_proj_qkv",
262
+ "model.language_model.layers.13.linear_attn.in_proj_z",
263
+ "model.language_model.layers.13.linear_attn.out_proj",
264
+ "model.language_model.layers.13.mlp.gate",
265
+ "model.language_model.layers.13.mlp.shared_expert.down_proj",
266
+ "model.language_model.layers.13.mlp.shared_expert.gate_proj",
267
+ "model.language_model.layers.13.mlp.shared_expert.up_proj",
268
+ "model.language_model.layers.13.mlp.shared_expert_gate",
269
+ "model.language_model.layers.13.mlp_hyper_connection.block_inject_weight",
270
+ "model.language_model.layers.13.mlp_hyper_connection.input_mix_weight_down",
271
+ "model.language_model.layers.13.mlp_hyper_connection.input_mix_weight_up",
272
+ "model.language_model.layers.14.attn_hyper_connection.block_inject_weight",
273
+ "model.language_model.layers.14.attn_hyper_connection.input_mix_weight_down",
274
+ "model.language_model.layers.14.attn_hyper_connection.input_mix_weight_up",
275
+ "model.language_model.layers.14.linear_attn.conv1d",
276
+ "model.language_model.layers.14.linear_attn.in_proj_a",
277
+ "model.language_model.layers.14.linear_attn.in_proj_b",
278
+ "model.language_model.layers.14.linear_attn.in_proj_qkv",
279
+ "model.language_model.layers.14.linear_attn.in_proj_z",
280
+ "model.language_model.layers.14.linear_attn.out_proj",
281
+ "model.language_model.layers.14.mlp.gate",
282
+ "model.language_model.layers.14.mlp.shared_expert.down_proj",
283
+ "model.language_model.layers.14.mlp.shared_expert.gate_proj",
284
+ "model.language_model.layers.14.mlp.shared_expert.up_proj",
285
+ "model.language_model.layers.14.mlp.shared_expert_gate",
286
+ "model.language_model.layers.14.mlp_hyper_connection.block_inject_weight",
287
+ "model.language_model.layers.14.mlp_hyper_connection.input_mix_weight_down",
288
+ "model.language_model.layers.14.mlp_hyper_connection.input_mix_weight_up",
289
+ "model.language_model.layers.15.attn_hyper_connection.block_inject_weight",
290
+ "model.language_model.layers.15.attn_hyper_connection.input_mix_weight_down",
291
+ "model.language_model.layers.15.attn_hyper_connection.input_mix_weight_up",
292
+ "model.language_model.layers.15.mlp.gate",
293
+ "model.language_model.layers.15.mlp.shared_expert.down_proj",
294
+ "model.language_model.layers.15.mlp.shared_expert.gate_proj",
295
+ "model.language_model.layers.15.mlp.shared_expert.up_proj",
296
+ "model.language_model.layers.15.mlp.shared_expert_gate",
297
+ "model.language_model.layers.15.mlp_hyper_connection.block_inject_weight",
298
+ "model.language_model.layers.15.mlp_hyper_connection.input_mix_weight_down",
299
+ "model.language_model.layers.15.mlp_hyper_connection.input_mix_weight_up",
300
+ "model.language_model.layers.15.self_attn.indexer.index_qk_proj",
301
+ "model.language_model.layers.15.self_attn.k_proj",
302
+ "model.language_model.layers.15.self_attn.o_proj",
303
+ "model.language_model.layers.15.self_attn.q_proj",
304
+ "model.language_model.layers.15.self_attn.v_proj",
305
+ "model.language_model.layers.16.attn_hyper_connection.block_inject_weight",
306
+ "model.language_model.layers.16.attn_hyper_connection.input_mix_weight_down",
307
+ "model.language_model.layers.16.attn_hyper_connection.input_mix_weight_up",
308
+ "model.language_model.layers.16.linear_attn.conv1d",
309
+ "model.language_model.layers.16.linear_attn.in_proj_a",
310
+ "model.language_model.layers.16.linear_attn.in_proj_b",
311
+ "model.language_model.layers.16.linear_attn.in_proj_qkv",
312
+ "model.language_model.layers.16.linear_attn.in_proj_z",
313
+ "model.language_model.layers.16.linear_attn.out_proj",
314
+ "model.language_model.layers.16.mlp.gate",
315
+ "model.language_model.layers.16.mlp.shared_expert.down_proj",
316
+ "model.language_model.layers.16.mlp.shared_expert.gate_proj",
317
+ "model.language_model.layers.16.mlp.shared_expert.up_proj",
318
+ "model.language_model.layers.16.mlp.shared_expert_gate",
319
+ "model.language_model.layers.16.mlp_hyper_connection.block_inject_weight",
320
+ "model.language_model.layers.16.mlp_hyper_connection.input_mix_weight_down",
321
+ "model.language_model.layers.16.mlp_hyper_connection.input_mix_weight_up",
322
+ "model.language_model.layers.17.attn_hyper_connection.block_inject_weight",
323
+ "model.language_model.layers.17.attn_hyper_connection.input_mix_weight_down",
324
+ "model.language_model.layers.17.attn_hyper_connection.input_mix_weight_up",
325
+ "model.language_model.layers.17.linear_attn.conv1d",
326
+ "model.language_model.layers.17.linear_attn.in_proj_a",
327
+ "model.language_model.layers.17.linear_attn.in_proj_b",
328
+ "model.language_model.layers.17.linear_attn.in_proj_qkv",
329
+ "model.language_model.layers.17.linear_attn.in_proj_z",
330
+ "model.language_model.layers.17.linear_attn.out_proj",
331
+ "model.language_model.layers.17.mlp.gate",
332
+ "model.language_model.layers.17.mlp.shared_expert.down_proj",
333
+ "model.language_model.layers.17.mlp.shared_expert.gate_proj",
334
+ "model.language_model.layers.17.mlp.shared_expert.up_proj",
335
+ "model.language_model.layers.17.mlp.shared_expert_gate",
336
+ "model.language_model.layers.17.mlp_hyper_connection.block_inject_weight",
337
+ "model.language_model.layers.17.mlp_hyper_connection.input_mix_weight_down",
338
+ "model.language_model.layers.17.mlp_hyper_connection.input_mix_weight_up",
339
+ "model.language_model.layers.18.attn_hyper_connection.block_inject_weight",
340
+ "model.language_model.layers.18.attn_hyper_connection.input_mix_weight_down",
341
+ "model.language_model.layers.18.attn_hyper_connection.input_mix_weight_up",
342
+ "model.language_model.layers.18.linear_attn.conv1d",
343
+ "model.language_model.layers.18.linear_attn.in_proj_a",
344
+ "model.language_model.layers.18.linear_attn.in_proj_b",
345
+ "model.language_model.layers.18.linear_attn.in_proj_qkv",
346
+ "model.language_model.layers.18.linear_attn.in_proj_z",
347
+ "model.language_model.layers.18.linear_attn.out_proj",
348
+ "model.language_model.layers.18.mlp.gate",
349
+ "model.language_model.layers.18.mlp.shared_expert.down_proj",
350
+ "model.language_model.layers.18.mlp.shared_expert.gate_proj",
351
+ "model.language_model.layers.18.mlp.shared_expert.up_proj",
352
+ "model.language_model.layers.18.mlp.shared_expert_gate",
353
+ "model.language_model.layers.18.mlp_hyper_connection.block_inject_weight",
354
+ "model.language_model.layers.18.mlp_hyper_connection.input_mix_weight_down",
355
+ "model.language_model.layers.18.mlp_hyper_connection.input_mix_weight_up",
356
+ "model.language_model.layers.19.attn_hyper_connection.block_inject_weight",
357
+ "model.language_model.layers.19.attn_hyper_connection.input_mix_weight_down",
358
+ "model.language_model.layers.19.attn_hyper_connection.input_mix_weight_up",
359
+ "model.language_model.layers.19.mlp.gate",
360
+ "model.language_model.layers.19.mlp.shared_expert.down_proj",
361
+ "model.language_model.layers.19.mlp.shared_expert.gate_proj",
362
+ "model.language_model.layers.19.mlp.shared_expert.up_proj",
363
+ "model.language_model.layers.19.mlp.shared_expert_gate",
364
+ "model.language_model.layers.19.mlp_hyper_connection.block_inject_weight",
365
+ "model.language_model.layers.19.mlp_hyper_connection.input_mix_weight_down",
366
+ "model.language_model.layers.19.mlp_hyper_connection.input_mix_weight_up",
367
+ "model.language_model.layers.19.self_attn.indexer.index_qk_proj",
368
+ "model.language_model.layers.19.self_attn.k_proj",
369
+ "model.language_model.layers.19.self_attn.o_proj",
370
+ "model.language_model.layers.19.self_attn.q_proj",
371
+ "model.language_model.layers.19.self_attn.v_proj",
372
+ "model.language_model.layers.2.attn_hyper_connection.block_inject_weight",
373
+ "model.language_model.layers.2.attn_hyper_connection.input_mix_weight_down",
374
+ "model.language_model.layers.2.attn_hyper_connection.input_mix_weight_up",
375
+ "model.language_model.layers.2.linear_attn.conv1d",
376
+ "model.language_model.layers.2.linear_attn.in_proj_a",
377
+ "model.language_model.layers.2.linear_attn.in_proj_b",
378
+ "model.language_model.layers.2.linear_attn.in_proj_qkv",
379
+ "model.language_model.layers.2.linear_attn.in_proj_z",
380
+ "model.language_model.layers.2.linear_attn.out_proj",
381
+ "model.language_model.layers.2.mlp.gate",
382
+ "model.language_model.layers.2.mlp.shared_expert.down_proj",
383
+ "model.language_model.layers.2.mlp.shared_expert.gate_proj",
384
+ "model.language_model.layers.2.mlp.shared_expert.up_proj",
385
+ "model.language_model.layers.2.mlp.shared_expert_gate",
386
+ "model.language_model.layers.2.mlp_hyper_connection.block_inject_weight",
387
+ "model.language_model.layers.2.mlp_hyper_connection.input_mix_weight_down",
388
+ "model.language_model.layers.2.mlp_hyper_connection.input_mix_weight_up",
389
+ "model.language_model.layers.20.attn_hyper_connection.block_inject_weight",
390
+ "model.language_model.layers.20.attn_hyper_connection.input_mix_weight_down",
391
+ "model.language_model.layers.20.attn_hyper_connection.input_mix_weight_up",
392
+ "model.language_model.layers.20.linear_attn.conv1d",
393
+ "model.language_model.layers.20.linear_attn.in_proj_a",
394
+ "model.language_model.layers.20.linear_attn.in_proj_b",
395
+ "model.language_model.layers.20.linear_attn.in_proj_qkv",
396
+ "model.language_model.layers.20.linear_attn.in_proj_z",
397
+ "model.language_model.layers.20.linear_attn.out_proj",
398
+ "model.language_model.layers.20.mlp.gate",
399
+ "model.language_model.layers.20.mlp.shared_expert.down_proj",
400
+ "model.language_model.layers.20.mlp.shared_expert.gate_proj",
401
+ "model.language_model.layers.20.mlp.shared_expert.up_proj",
402
+ "model.language_model.layers.20.mlp.shared_expert_gate",
403
+ "model.language_model.layers.20.mlp_hyper_connection.block_inject_weight",
404
+ "model.language_model.layers.20.mlp_hyper_connection.input_mix_weight_down",
405
+ "model.language_model.layers.20.mlp_hyper_connection.input_mix_weight_up",
406
+ "model.language_model.layers.21.attn_hyper_connection.block_inject_weight",
407
+ "model.language_model.layers.21.attn_hyper_connection.input_mix_weight_down",
408
+ "model.language_model.layers.21.attn_hyper_connection.input_mix_weight_up",
409
+ "model.language_model.layers.21.linear_attn.conv1d",
410
+ "model.language_model.layers.21.linear_attn.in_proj_a",
411
+ "model.language_model.layers.21.linear_attn.in_proj_b",
412
+ "model.language_model.layers.21.linear_attn.in_proj_qkv",
413
+ "model.language_model.layers.21.linear_attn.in_proj_z",
414
+ "model.language_model.layers.21.linear_attn.out_proj",
415
+ "model.language_model.layers.21.mlp.gate",
416
+ "model.language_model.layers.21.mlp.shared_expert.down_proj",
417
+ "model.language_model.layers.21.mlp.shared_expert.gate_proj",
418
+ "model.language_model.layers.21.mlp.shared_expert.up_proj",
419
+ "model.language_model.layers.21.mlp.shared_expert_gate",
420
+ "model.language_model.layers.21.mlp_hyper_connection.block_inject_weight",
421
+ "model.language_model.layers.21.mlp_hyper_connection.input_mix_weight_down",
422
+ "model.language_model.layers.21.mlp_hyper_connection.input_mix_weight_up",
423
+ "model.language_model.layers.22.attn_hyper_connection.block_inject_weight",
424
+ "model.language_model.layers.22.attn_hyper_connection.input_mix_weight_down",
425
+ "model.language_model.layers.22.attn_hyper_connection.input_mix_weight_up",
426
+ "model.language_model.layers.22.linear_attn.conv1d",
427
+ "model.language_model.layers.22.linear_attn.in_proj_a",
428
+ "model.language_model.layers.22.linear_attn.in_proj_b",
429
+ "model.language_model.layers.22.linear_attn.in_proj_qkv",
430
+ "model.language_model.layers.22.linear_attn.in_proj_z",
431
+ "model.language_model.layers.22.linear_attn.out_proj",
432
+ "model.language_model.layers.22.mlp.gate",
433
+ "model.language_model.layers.22.mlp.shared_expert.down_proj",
434
+ "model.language_model.layers.22.mlp.shared_expert.gate_proj",
435
+ "model.language_model.layers.22.mlp.shared_expert.up_proj",
436
+ "model.language_model.layers.22.mlp.shared_expert_gate",
437
+ "model.language_model.layers.22.mlp_hyper_connection.block_inject_weight",
438
+ "model.language_model.layers.22.mlp_hyper_connection.input_mix_weight_down",
439
+ "model.language_model.layers.22.mlp_hyper_connection.input_mix_weight_up",
440
+ "model.language_model.layers.23.attn_hyper_connection.block_inject_weight",
441
+ "model.language_model.layers.23.attn_hyper_connection.input_mix_weight_down",
442
+ "model.language_model.layers.23.attn_hyper_connection.input_mix_weight_up",
443
+ "model.language_model.layers.23.mlp.gate",
444
+ "model.language_model.layers.23.mlp.shared_expert.down_proj",
445
+ "model.language_model.layers.23.mlp.shared_expert.gate_proj",
446
+ "model.language_model.layers.23.mlp.shared_expert.up_proj",
447
+ "model.language_model.layers.23.mlp.shared_expert_gate",
448
+ "model.language_model.layers.23.mlp_hyper_connection.block_inject_weight",
449
+ "model.language_model.layers.23.mlp_hyper_connection.input_mix_weight_down",
450
+ "model.language_model.layers.23.mlp_hyper_connection.input_mix_weight_up",
451
+ "model.language_model.layers.23.self_attn.indexer.index_qk_proj",
452
+ "model.language_model.layers.23.self_attn.k_proj",
453
+ "model.language_model.layers.23.self_attn.o_proj",
454
+ "model.language_model.layers.23.self_attn.q_proj",
455
+ "model.language_model.layers.23.self_attn.v_proj",
456
+ "model.language_model.layers.24.attn_hyper_connection.block_inject_weight",
457
+ "model.language_model.layers.24.attn_hyper_connection.input_mix_weight_down",
458
+ "model.language_model.layers.24.attn_hyper_connection.input_mix_weight_up",
459
+ "model.language_model.layers.24.linear_attn.conv1d",
460
+ "model.language_model.layers.24.linear_attn.in_proj_a",
461
+ "model.language_model.layers.24.linear_attn.in_proj_b",
462
+ "model.language_model.layers.24.linear_attn.in_proj_qkv",
463
+ "model.language_model.layers.24.linear_attn.in_proj_z",
464
+ "model.language_model.layers.24.linear_attn.out_proj",
465
+ "model.language_model.layers.24.mlp.gate",
466
+ "model.language_model.layers.24.mlp.shared_expert.down_proj",
467
+ "model.language_model.layers.24.mlp.shared_expert.gate_proj",
468
+ "model.language_model.layers.24.mlp.shared_expert.up_proj",
469
+ "model.language_model.layers.24.mlp.shared_expert_gate",
470
+ "model.language_model.layers.24.mlp_hyper_connection.block_inject_weight",
471
+ "model.language_model.layers.24.mlp_hyper_connection.input_mix_weight_down",
472
+ "model.language_model.layers.24.mlp_hyper_connection.input_mix_weight_up",
473
+ "model.language_model.layers.25.attn_hyper_connection.block_inject_weight",
474
+ "model.language_model.layers.25.attn_hyper_connection.input_mix_weight_down",
475
+ "model.language_model.layers.25.attn_hyper_connection.input_mix_weight_up",
476
+ "model.language_model.layers.25.linear_attn.conv1d",
477
+ "model.language_model.layers.25.linear_attn.in_proj_a",
478
+ "model.language_model.layers.25.linear_attn.in_proj_b",
479
+ "model.language_model.layers.25.linear_attn.in_proj_qkv",
480
+ "model.language_model.layers.25.linear_attn.in_proj_z",
481
+ "model.language_model.layers.25.linear_attn.out_proj",
482
+ "model.language_model.layers.25.mlp.gate",
483
+ "model.language_model.layers.25.mlp.shared_expert.down_proj",
484
+ "model.language_model.layers.25.mlp.shared_expert.gate_proj",
485
+ "model.language_model.layers.25.mlp.shared_expert.up_proj",
486
+ "model.language_model.layers.25.mlp.shared_expert_gate",
487
+ "model.language_model.layers.25.mlp_hyper_connection.block_inject_weight",
488
+ "model.language_model.layers.25.mlp_hyper_connection.input_mix_weight_down",
489
+ "model.language_model.layers.25.mlp_hyper_connection.input_mix_weight_up",
490
+ "model.language_model.layers.26.attn_hyper_connection.block_inject_weight",
491
+ "model.language_model.layers.26.attn_hyper_connection.input_mix_weight_down",
492
+ "model.language_model.layers.26.attn_hyper_connection.input_mix_weight_up",
493
+ "model.language_model.layers.26.linear_attn.conv1d",
494
+ "model.language_model.layers.26.linear_attn.in_proj_a",
495
+ "model.language_model.layers.26.linear_attn.in_proj_b",
496
+ "model.language_model.layers.26.linear_attn.in_proj_qkv",
497
+ "model.language_model.layers.26.linear_attn.in_proj_z",
498
+ "model.language_model.layers.26.linear_attn.out_proj",
499
+ "model.language_model.layers.26.mlp.gate",
500
+ "model.language_model.layers.26.mlp.shared_expert.down_proj",
501
+ "model.language_model.layers.26.mlp.shared_expert.gate_proj",
502
+ "model.language_model.layers.26.mlp.shared_expert.up_proj",
503
+ "model.language_model.layers.26.mlp.shared_expert_gate",
504
+ "model.language_model.layers.26.mlp_hyper_connection.block_inject_weight",
505
+ "model.language_model.layers.26.mlp_hyper_connection.input_mix_weight_down",
506
+ "model.language_model.layers.26.mlp_hyper_connection.input_mix_weight_up",
507
+ "model.language_model.layers.27.attn_hyper_connection.block_inject_weight",
508
+ "model.language_model.layers.27.attn_hyper_connection.input_mix_weight_down",
509
+ "model.language_model.layers.27.attn_hyper_connection.input_mix_weight_up",
510
+ "model.language_model.layers.27.mlp.gate",
511
+ "model.language_model.layers.27.mlp.shared_expert.down_proj",
512
+ "model.language_model.layers.27.mlp.shared_expert.gate_proj",
513
+ "model.language_model.layers.27.mlp.shared_expert.up_proj",
514
+ "model.language_model.layers.27.mlp.shared_expert_gate",
515
+ "model.language_model.layers.27.mlp_hyper_connection.block_inject_weight",
516
+ "model.language_model.layers.27.mlp_hyper_connection.input_mix_weight_down",
517
+ "model.language_model.layers.27.mlp_hyper_connection.input_mix_weight_up",
518
+ "model.language_model.layers.27.self_attn.indexer.index_qk_proj",
519
+ "model.language_model.layers.27.self_attn.k_proj",
520
+ "model.language_model.layers.27.self_attn.o_proj",
521
+ "model.language_model.layers.27.self_attn.q_proj",
522
+ "model.language_model.layers.27.self_attn.v_proj",
523
+ "model.language_model.layers.28.attn_hyper_connection.block_inject_weight",
524
+ "model.language_model.layers.28.attn_hyper_connection.input_mix_weight_down",
525
+ "model.language_model.layers.28.attn_hyper_connection.input_mix_weight_up",
526
+ "model.language_model.layers.28.linear_attn.conv1d",
527
+ "model.language_model.layers.28.linear_attn.in_proj_a",
528
+ "model.language_model.layers.28.linear_attn.in_proj_b",
529
+ "model.language_model.layers.28.linear_attn.in_proj_qkv",
530
+ "model.language_model.layers.28.linear_attn.in_proj_z",
531
+ "model.language_model.layers.28.linear_attn.out_proj",
532
+ "model.language_model.layers.28.mlp.gate",
533
+ "model.language_model.layers.28.mlp.shared_expert.down_proj",
534
+ "model.language_model.layers.28.mlp.shared_expert.gate_proj",
535
+ "model.language_model.layers.28.mlp.shared_expert.up_proj",
536
+ "model.language_model.layers.28.mlp.shared_expert_gate",
537
+ "model.language_model.layers.28.mlp_hyper_connection.block_inject_weight",
538
+ "model.language_model.layers.28.mlp_hyper_connection.input_mix_weight_down",
539
+ "model.language_model.layers.28.mlp_hyper_connection.input_mix_weight_up",
540
+ "model.language_model.layers.29.attn_hyper_connection.block_inject_weight",
541
+ "model.language_model.layers.29.attn_hyper_connection.input_mix_weight_down",
542
+ "model.language_model.layers.29.attn_hyper_connection.input_mix_weight_up",
543
+ "model.language_model.layers.29.linear_attn.conv1d",
544
+ "model.language_model.layers.29.linear_attn.in_proj_a",
545
+ "model.language_model.layers.29.linear_attn.in_proj_b",
546
+ "model.language_model.layers.29.linear_attn.in_proj_qkv",
547
+ "model.language_model.layers.29.linear_attn.in_proj_z",
548
+ "model.language_model.layers.29.linear_attn.out_proj",
549
+ "model.language_model.layers.29.mlp.gate",
550
+ "model.language_model.layers.29.mlp.shared_expert.down_proj",
551
+ "model.language_model.layers.29.mlp.shared_expert.gate_proj",
552
+ "model.language_model.layers.29.mlp.shared_expert.up_proj",
553
+ "model.language_model.layers.29.mlp.shared_expert_gate",
554
+ "model.language_model.layers.29.mlp_hyper_connection.block_inject_weight",
555
+ "model.language_model.layers.29.mlp_hyper_connection.input_mix_weight_down",
556
+ "model.language_model.layers.29.mlp_hyper_connection.input_mix_weight_up",
557
+ "model.language_model.layers.3.attn_hyper_connection.block_inject_weight",
558
+ "model.language_model.layers.3.attn_hyper_connection.input_mix_weight_down",
559
+ "model.language_model.layers.3.attn_hyper_connection.input_mix_weight_up",
560
+ "model.language_model.layers.3.mlp.gate",
561
+ "model.language_model.layers.3.mlp.shared_expert.down_proj",
562
+ "model.language_model.layers.3.mlp.shared_expert.gate_proj",
563
+ "model.language_model.layers.3.mlp.shared_expert.up_proj",
564
+ "model.language_model.layers.3.mlp.shared_expert_gate",
565
+ "model.language_model.layers.3.mlp_hyper_connection.block_inject_weight",
566
+ "model.language_model.layers.3.mlp_hyper_connection.input_mix_weight_down",
567
+ "model.language_model.layers.3.mlp_hyper_connection.input_mix_weight_up",
568
+ "model.language_model.layers.3.self_attn.indexer.index_qk_proj",
569
+ "model.language_model.layers.3.self_attn.k_proj",
570
+ "model.language_model.layers.3.self_attn.o_proj",
571
+ "model.language_model.layers.3.self_attn.q_proj",
572
+ "model.language_model.layers.3.self_attn.v_proj",
573
+ "model.language_model.layers.30.attn_hyper_connection.block_inject_weight",
574
+ "model.language_model.layers.30.attn_hyper_connection.input_mix_weight_down",
575
+ "model.language_model.layers.30.attn_hyper_connection.input_mix_weight_up",
576
+ "model.language_model.layers.30.linear_attn.conv1d",
577
+ "model.language_model.layers.30.linear_attn.in_proj_a",
578
+ "model.language_model.layers.30.linear_attn.in_proj_b",
579
+ "model.language_model.layers.30.linear_attn.in_proj_qkv",
580
+ "model.language_model.layers.30.linear_attn.in_proj_z",
581
+ "model.language_model.layers.30.linear_attn.out_proj",
582
+ "model.language_model.layers.30.mlp.gate",
583
+ "model.language_model.layers.30.mlp.shared_expert.down_proj",
584
+ "model.language_model.layers.30.mlp.shared_expert.gate_proj",
585
+ "model.language_model.layers.30.mlp.shared_expert.up_proj",
586
+ "model.language_model.layers.30.mlp.shared_expert_gate",
587
+ "model.language_model.layers.30.mlp_hyper_connection.block_inject_weight",
588
+ "model.language_model.layers.30.mlp_hyper_connection.input_mix_weight_down",
589
+ "model.language_model.layers.30.mlp_hyper_connection.input_mix_weight_up",
590
+ "model.language_model.layers.31.attn_hyper_connection.block_inject_weight",
591
+ "model.language_model.layers.31.attn_hyper_connection.input_mix_weight_down",
592
+ "model.language_model.layers.31.attn_hyper_connection.input_mix_weight_up",
593
+ "model.language_model.layers.31.mlp.gate",
594
+ "model.language_model.layers.31.mlp.shared_expert.down_proj",
595
+ "model.language_model.layers.31.mlp.shared_expert.gate_proj",
596
+ "model.language_model.layers.31.mlp.shared_expert.up_proj",
597
+ "model.language_model.layers.31.mlp.shared_expert_gate",
598
+ "model.language_model.layers.31.mlp_hyper_connection.block_inject_weight",
599
+ "model.language_model.layers.31.mlp_hyper_connection.input_mix_weight_down",
600
+ "model.language_model.layers.31.mlp_hyper_connection.input_mix_weight_up",
601
+ "model.language_model.layers.31.self_attn.indexer.index_qk_proj",
602
+ "model.language_model.layers.31.self_attn.k_proj",
603
+ "model.language_model.layers.31.self_attn.o_proj",
604
+ "model.language_model.layers.31.self_attn.q_proj",
605
+ "model.language_model.layers.31.self_attn.v_proj",
606
+ "model.language_model.layers.32.attn_hyper_connection.block_inject_weight",
607
+ "model.language_model.layers.32.attn_hyper_connection.input_mix_weight_down",
608
+ "model.language_model.layers.32.attn_hyper_connection.input_mix_weight_up",
609
+ "model.language_model.layers.32.linear_attn.conv1d",
610
+ "model.language_model.layers.32.linear_attn.in_proj_a",
611
+ "model.language_model.layers.32.linear_attn.in_proj_b",
612
+ "model.language_model.layers.32.linear_attn.in_proj_qkv",
613
+ "model.language_model.layers.32.linear_attn.in_proj_z",
614
+ "model.language_model.layers.32.linear_attn.out_proj",
615
+ "model.language_model.layers.32.mlp.gate",
616
+ "model.language_model.layers.32.mlp.shared_expert.down_proj",
617
+ "model.language_model.layers.32.mlp.shared_expert.gate_proj",
618
+ "model.language_model.layers.32.mlp.shared_expert.up_proj",
619
+ "model.language_model.layers.32.mlp.shared_expert_gate",
620
+ "model.language_model.layers.32.mlp_hyper_connection.block_inject_weight",
621
+ "model.language_model.layers.32.mlp_hyper_connection.input_mix_weight_down",
622
+ "model.language_model.layers.32.mlp_hyper_connection.input_mix_weight_up",
623
+ "model.language_model.layers.33.attn_hyper_connection.block_inject_weight",
624
+ "model.language_model.layers.33.attn_hyper_connection.input_mix_weight_down",
625
+ "model.language_model.layers.33.attn_hyper_connection.input_mix_weight_up",
626
+ "model.language_model.layers.33.linear_attn.conv1d",
627
+ "model.language_model.layers.33.linear_attn.in_proj_a",
628
+ "model.language_model.layers.33.linear_attn.in_proj_b",
629
+ "model.language_model.layers.33.linear_attn.in_proj_qkv",
630
+ "model.language_model.layers.33.linear_attn.in_proj_z",
631
+ "model.language_model.layers.33.linear_attn.out_proj",
632
+ "model.language_model.layers.33.mlp.gate",
633
+ "model.language_model.layers.33.mlp.shared_expert.down_proj",
634
+ "model.language_model.layers.33.mlp.shared_expert.gate_proj",
635
+ "model.language_model.layers.33.mlp.shared_expert.up_proj",
636
+ "model.language_model.layers.33.mlp.shared_expert_gate",
637
+ "model.language_model.layers.33.mlp_hyper_connection.block_inject_weight",
638
+ "model.language_model.layers.33.mlp_hyper_connection.input_mix_weight_down",
639
+ "model.language_model.layers.33.mlp_hyper_connection.input_mix_weight_up",
640
+ "model.language_model.layers.34.attn_hyper_connection.block_inject_weight",
641
+ "model.language_model.layers.34.attn_hyper_connection.input_mix_weight_down",
642
+ "model.language_model.layers.34.attn_hyper_connection.input_mix_weight_up",
643
+ "model.language_model.layers.34.linear_attn.conv1d",
644
+ "model.language_model.layers.34.linear_attn.in_proj_a",
645
+ "model.language_model.layers.34.linear_attn.in_proj_b",
646
+ "model.language_model.layers.34.linear_attn.in_proj_qkv",
647
+ "model.language_model.layers.34.linear_attn.in_proj_z",
648
+ "model.language_model.layers.34.linear_attn.out_proj",
649
+ "model.language_model.layers.34.mlp.gate",
650
+ "model.language_model.layers.34.mlp.shared_expert.down_proj",
651
+ "model.language_model.layers.34.mlp.shared_expert.gate_proj",
652
+ "model.language_model.layers.34.mlp.shared_expert.up_proj",
653
+ "model.language_model.layers.34.mlp.shared_expert_gate",
654
+ "model.language_model.layers.34.mlp_hyper_connection.block_inject_weight",
655
+ "model.language_model.layers.34.mlp_hyper_connection.input_mix_weight_down",
656
+ "model.language_model.layers.34.mlp_hyper_connection.input_mix_weight_up",
657
+ "model.language_model.layers.35.attn_hyper_connection.block_inject_weight",
658
+ "model.language_model.layers.35.attn_hyper_connection.input_mix_weight_down",
659
+ "model.language_model.layers.35.attn_hyper_connection.input_mix_weight_up",
660
+ "model.language_model.layers.35.mlp.gate",
661
+ "model.language_model.layers.35.mlp.shared_expert.down_proj",
662
+ "model.language_model.layers.35.mlp.shared_expert.gate_proj",
663
+ "model.language_model.layers.35.mlp.shared_expert.up_proj",
664
+ "model.language_model.layers.35.mlp.shared_expert_gate",
665
+ "model.language_model.layers.35.mlp_hyper_connection.block_inject_weight",
666
+ "model.language_model.layers.35.mlp_hyper_connection.input_mix_weight_down",
667
+ "model.language_model.layers.35.mlp_hyper_connection.input_mix_weight_up",
668
+ "model.language_model.layers.35.self_attn.indexer.index_qk_proj",
669
+ "model.language_model.layers.35.self_attn.k_proj",
670
+ "model.language_model.layers.35.self_attn.o_proj",
671
+ "model.language_model.layers.35.self_attn.q_proj",
672
+ "model.language_model.layers.35.self_attn.v_proj",
673
+ "model.language_model.layers.36.attn_hyper_connection.block_inject_weight",
674
+ "model.language_model.layers.36.attn_hyper_connection.input_mix_weight_down",
675
+ "model.language_model.layers.36.attn_hyper_connection.input_mix_weight_up",
676
+ "model.language_model.layers.36.linear_attn.conv1d",
677
+ "model.language_model.layers.36.linear_attn.in_proj_a",
678
+ "model.language_model.layers.36.linear_attn.in_proj_b",
679
+ "model.language_model.layers.36.linear_attn.in_proj_qkv",
680
+ "model.language_model.layers.36.linear_attn.in_proj_z",
681
+ "model.language_model.layers.36.linear_attn.out_proj",
682
+ "model.language_model.layers.36.mlp.gate",
683
+ "model.language_model.layers.36.mlp.shared_expert.down_proj",
684
+ "model.language_model.layers.36.mlp.shared_expert.gate_proj",
685
+ "model.language_model.layers.36.mlp.shared_expert.up_proj",
686
+ "model.language_model.layers.36.mlp.shared_expert_gate",
687
+ "model.language_model.layers.36.mlp_hyper_connection.block_inject_weight",
688
+ "model.language_model.layers.36.mlp_hyper_connection.input_mix_weight_down",
689
+ "model.language_model.layers.36.mlp_hyper_connection.input_mix_weight_up",
690
+ "model.language_model.layers.37.attn_hyper_connection.block_inject_weight",
691
+ "model.language_model.layers.37.attn_hyper_connection.input_mix_weight_down",
692
+ "model.language_model.layers.37.attn_hyper_connection.input_mix_weight_up",
693
+ "model.language_model.layers.37.linear_attn.conv1d",
694
+ "model.language_model.layers.37.linear_attn.in_proj_a",
695
+ "model.language_model.layers.37.linear_attn.in_proj_b",
696
+ "model.language_model.layers.37.linear_attn.in_proj_qkv",
697
+ "model.language_model.layers.37.linear_attn.in_proj_z",
698
+ "model.language_model.layers.37.linear_attn.out_proj",
699
+ "model.language_model.layers.37.mlp.gate",
700
+ "model.language_model.layers.37.mlp.shared_expert.down_proj",
701
+ "model.language_model.layers.37.mlp.shared_expert.gate_proj",
702
+ "model.language_model.layers.37.mlp.shared_expert.up_proj",
703
+ "model.language_model.layers.37.mlp.shared_expert_gate",
704
+ "model.language_model.layers.37.mlp_hyper_connection.block_inject_weight",
705
+ "model.language_model.layers.37.mlp_hyper_connection.input_mix_weight_down",
706
+ "model.language_model.layers.37.mlp_hyper_connection.input_mix_weight_up",
707
+ "model.language_model.layers.38.attn_hyper_connection.block_inject_weight",
708
+ "model.language_model.layers.38.attn_hyper_connection.input_mix_weight_down",
709
+ "model.language_model.layers.38.attn_hyper_connection.input_mix_weight_up",
710
+ "model.language_model.layers.38.linear_attn.conv1d",
711
+ "model.language_model.layers.38.linear_attn.in_proj_a",
712
+ "model.language_model.layers.38.linear_attn.in_proj_b",
713
+ "model.language_model.layers.38.linear_attn.in_proj_qkv",
714
+ "model.language_model.layers.38.linear_attn.in_proj_z",
715
+ "model.language_model.layers.38.linear_attn.out_proj",
716
+ "model.language_model.layers.38.mlp.gate",
717
+ "model.language_model.layers.38.mlp.shared_expert.down_proj",
718
+ "model.language_model.layers.38.mlp.shared_expert.gate_proj",
719
+ "model.language_model.layers.38.mlp.shared_expert.up_proj",
720
+ "model.language_model.layers.38.mlp.shared_expert_gate",
721
+ "model.language_model.layers.38.mlp_hyper_connection.block_inject_weight",
722
+ "model.language_model.layers.38.mlp_hyper_connection.input_mix_weight_down",
723
+ "model.language_model.layers.38.mlp_hyper_connection.input_mix_weight_up",
724
+ "model.language_model.layers.39.attn_hyper_connection.block_inject_weight",
725
+ "model.language_model.layers.39.attn_hyper_connection.input_mix_weight_down",
726
+ "model.language_model.layers.39.attn_hyper_connection.input_mix_weight_up",
727
+ "model.language_model.layers.39.mlp.gate",
728
+ "model.language_model.layers.39.mlp.shared_expert.down_proj",
729
+ "model.language_model.layers.39.mlp.shared_expert.gate_proj",
730
+ "model.language_model.layers.39.mlp.shared_expert.up_proj",
731
+ "model.language_model.layers.39.mlp.shared_expert_gate",
732
+ "model.language_model.layers.39.mlp_hyper_connection.block_inject_weight",
733
+ "model.language_model.layers.39.mlp_hyper_connection.input_mix_weight_down",
734
+ "model.language_model.layers.39.mlp_hyper_connection.input_mix_weight_up",
735
+ "model.language_model.layers.39.self_attn.indexer.index_qk_proj",
736
+ "model.language_model.layers.39.self_attn.k_proj",
737
+ "model.language_model.layers.39.self_attn.o_proj",
738
+ "model.language_model.layers.39.self_attn.q_proj",
739
+ "model.language_model.layers.39.self_attn.v_proj",
740
+ "model.language_model.layers.4.attn_hyper_connection.block_inject_weight",
741
+ "model.language_model.layers.4.attn_hyper_connection.input_mix_weight_down",
742
+ "model.language_model.layers.4.attn_hyper_connection.input_mix_weight_up",
743
+ "model.language_model.layers.4.linear_attn.conv1d",
744
+ "model.language_model.layers.4.linear_attn.in_proj_a",
745
+ "model.language_model.layers.4.linear_attn.in_proj_b",
746
+ "model.language_model.layers.4.linear_attn.in_proj_qkv",
747
+ "model.language_model.layers.4.linear_attn.in_proj_z",
748
+ "model.language_model.layers.4.linear_attn.out_proj",
749
+ "model.language_model.layers.4.mlp.gate",
750
+ "model.language_model.layers.4.mlp.shared_expert.down_proj",
751
+ "model.language_model.layers.4.mlp.shared_expert.gate_proj",
752
+ "model.language_model.layers.4.mlp.shared_expert.up_proj",
753
+ "model.language_model.layers.4.mlp.shared_expert_gate",
754
+ "model.language_model.layers.4.mlp_hyper_connection.block_inject_weight",
755
+ "model.language_model.layers.4.mlp_hyper_connection.input_mix_weight_down",
756
+ "model.language_model.layers.4.mlp_hyper_connection.input_mix_weight_up",
757
+ "model.language_model.layers.40.attn_hyper_connection.block_inject_weight",
758
+ "model.language_model.layers.40.attn_hyper_connection.input_mix_weight_down",
759
+ "model.language_model.layers.40.attn_hyper_connection.input_mix_weight_up",
760
+ "model.language_model.layers.40.linear_attn.conv1d",
761
+ "model.language_model.layers.40.linear_attn.in_proj_a",
762
+ "model.language_model.layers.40.linear_attn.in_proj_b",
763
+ "model.language_model.layers.40.linear_attn.in_proj_qkv",
764
+ "model.language_model.layers.40.linear_attn.in_proj_z",
765
+ "model.language_model.layers.40.linear_attn.out_proj",
766
+ "model.language_model.layers.40.mlp.gate",
767
+ "model.language_model.layers.40.mlp.shared_expert.down_proj",
768
+ "model.language_model.layers.40.mlp.shared_expert.gate_proj",
769
+ "model.language_model.layers.40.mlp.shared_expert.up_proj",
770
+ "model.language_model.layers.40.mlp.shared_expert_gate",
771
+ "model.language_model.layers.40.mlp_hyper_connection.block_inject_weight",
772
+ "model.language_model.layers.40.mlp_hyper_connection.input_mix_weight_down",
773
+ "model.language_model.layers.40.mlp_hyper_connection.input_mix_weight_up",
774
+ "model.language_model.layers.41.attn_hyper_connection.block_inject_weight",
775
+ "model.language_model.layers.41.attn_hyper_connection.input_mix_weight_down",
776
+ "model.language_model.layers.41.attn_hyper_connection.input_mix_weight_up",
777
+ "model.language_model.layers.41.linear_attn.conv1d",
778
+ "model.language_model.layers.41.linear_attn.in_proj_a",
779
+ "model.language_model.layers.41.linear_attn.in_proj_b",
780
+ "model.language_model.layers.41.linear_attn.in_proj_qkv",
781
+ "model.language_model.layers.41.linear_attn.in_proj_z",
782
+ "model.language_model.layers.41.linear_attn.out_proj",
783
+ "model.language_model.layers.41.mlp.gate",
784
+ "model.language_model.layers.41.mlp.shared_expert.down_proj",
785
+ "model.language_model.layers.41.mlp.shared_expert.gate_proj",
786
+ "model.language_model.layers.41.mlp.shared_expert.up_proj",
787
+ "model.language_model.layers.41.mlp.shared_expert_gate",
788
+ "model.language_model.layers.41.mlp_hyper_connection.block_inject_weight",
789
+ "model.language_model.layers.41.mlp_hyper_connection.input_mix_weight_down",
790
+ "model.language_model.layers.41.mlp_hyper_connection.input_mix_weight_up",
791
+ "model.language_model.layers.42.attn_hyper_connection.block_inject_weight",
792
+ "model.language_model.layers.42.attn_hyper_connection.input_mix_weight_down",
793
+ "model.language_model.layers.42.attn_hyper_connection.input_mix_weight_up",
794
+ "model.language_model.layers.42.linear_attn.conv1d",
795
+ "model.language_model.layers.42.linear_attn.in_proj_a",
796
+ "model.language_model.layers.42.linear_attn.in_proj_b",
797
+ "model.language_model.layers.42.linear_attn.in_proj_qkv",
798
+ "model.language_model.layers.42.linear_attn.in_proj_z",
799
+ "model.language_model.layers.42.linear_attn.out_proj",
800
+ "model.language_model.layers.42.mlp.gate",
801
+ "model.language_model.layers.42.mlp.shared_expert.down_proj",
802
+ "model.language_model.layers.42.mlp.shared_expert.gate_proj",
803
+ "model.language_model.layers.42.mlp.shared_expert.up_proj",
804
+ "model.language_model.layers.42.mlp.shared_expert_gate",
805
+ "model.language_model.layers.42.mlp_hyper_connection.block_inject_weight",
806
+ "model.language_model.layers.42.mlp_hyper_connection.input_mix_weight_down",
807
+ "model.language_model.layers.42.mlp_hyper_connection.input_mix_weight_up",
808
+ "model.language_model.layers.43.attn_hyper_connection.block_inject_weight",
809
+ "model.language_model.layers.43.attn_hyper_connection.input_mix_weight_down",
810
+ "model.language_model.layers.43.attn_hyper_connection.input_mix_weight_up",
811
+ "model.language_model.layers.43.mlp.gate",
812
+ "model.language_model.layers.43.mlp.shared_expert.down_proj",
813
+ "model.language_model.layers.43.mlp.shared_expert.gate_proj",
814
+ "model.language_model.layers.43.mlp.shared_expert.up_proj",
815
+ "model.language_model.layers.43.mlp.shared_expert_gate",
816
+ "model.language_model.layers.43.mlp_hyper_connection.block_inject_weight",
817
+ "model.language_model.layers.43.mlp_hyper_connection.input_mix_weight_down",
818
+ "model.language_model.layers.43.mlp_hyper_connection.input_mix_weight_up",
819
+ "model.language_model.layers.43.self_attn.indexer.index_qk_proj",
820
+ "model.language_model.layers.43.self_attn.k_proj",
821
+ "model.language_model.layers.43.self_attn.o_proj",
822
+ "model.language_model.layers.43.self_attn.q_proj",
823
+ "model.language_model.layers.43.self_attn.v_proj",
824
+ "model.language_model.layers.44.attn_hyper_connection.block_inject_weight",
825
+ "model.language_model.layers.44.attn_hyper_connection.input_mix_weight_down",
826
+ "model.language_model.layers.44.attn_hyper_connection.input_mix_weight_up",
827
+ "model.language_model.layers.44.linear_attn.conv1d",
828
+ "model.language_model.layers.44.linear_attn.in_proj_a",
829
+ "model.language_model.layers.44.linear_attn.in_proj_b",
830
+ "model.language_model.layers.44.linear_attn.in_proj_qkv",
831
+ "model.language_model.layers.44.linear_attn.in_proj_z",
832
+ "model.language_model.layers.44.linear_attn.out_proj",
833
+ "model.language_model.layers.44.mlp.gate",
834
+ "model.language_model.layers.44.mlp.shared_expert.down_proj",
835
+ "model.language_model.layers.44.mlp.shared_expert.gate_proj",
836
+ "model.language_model.layers.44.mlp.shared_expert.up_proj",
837
+ "model.language_model.layers.44.mlp.shared_expert_gate",
838
+ "model.language_model.layers.44.mlp_hyper_connection.block_inject_weight",
839
+ "model.language_model.layers.44.mlp_hyper_connection.input_mix_weight_down",
840
+ "model.language_model.layers.44.mlp_hyper_connection.input_mix_weight_up",
841
+ "model.language_model.layers.45.attn_hyper_connection.block_inject_weight",
842
+ "model.language_model.layers.45.attn_hyper_connection.input_mix_weight_down",
843
+ "model.language_model.layers.45.attn_hyper_connection.input_mix_weight_up",
844
+ "model.language_model.layers.45.linear_attn.conv1d",
845
+ "model.language_model.layers.45.linear_attn.in_proj_a",
846
+ "model.language_model.layers.45.linear_attn.in_proj_b",
847
+ "model.language_model.layers.45.linear_attn.in_proj_qkv",
848
+ "model.language_model.layers.45.linear_attn.in_proj_z",
849
+ "model.language_model.layers.45.linear_attn.out_proj",
850
+ "model.language_model.layers.45.mlp.gate",
851
+ "model.language_model.layers.45.mlp.shared_expert.down_proj",
852
+ "model.language_model.layers.45.mlp.shared_expert.gate_proj",
853
+ "model.language_model.layers.45.mlp.shared_expert.up_proj",
854
+ "model.language_model.layers.45.mlp.shared_expert_gate",
855
+ "model.language_model.layers.45.mlp_hyper_connection.block_inject_weight",
856
+ "model.language_model.layers.45.mlp_hyper_connection.input_mix_weight_down",
857
+ "model.language_model.layers.45.mlp_hyper_connection.input_mix_weight_up",
858
+ "model.language_model.layers.46.attn_hyper_connection.block_inject_weight",
859
+ "model.language_model.layers.46.attn_hyper_connection.input_mix_weight_down",
860
+ "model.language_model.layers.46.attn_hyper_connection.input_mix_weight_up",
861
+ "model.language_model.layers.46.linear_attn.conv1d",
862
+ "model.language_model.layers.46.linear_attn.in_proj_a",
863
+ "model.language_model.layers.46.linear_attn.in_proj_b",
864
+ "model.language_model.layers.46.linear_attn.in_proj_qkv",
865
+ "model.language_model.layers.46.linear_attn.in_proj_z",
866
+ "model.language_model.layers.46.linear_attn.out_proj",
867
+ "model.language_model.layers.46.mlp.gate",
868
+ "model.language_model.layers.46.mlp.shared_expert.down_proj",
869
+ "model.language_model.layers.46.mlp.shared_expert.gate_proj",
870
+ "model.language_model.layers.46.mlp.shared_expert.up_proj",
871
+ "model.language_model.layers.46.mlp.shared_expert_gate",
872
+ "model.language_model.layers.46.mlp_hyper_connection.block_inject_weight",
873
+ "model.language_model.layers.46.mlp_hyper_connection.input_mix_weight_down",
874
+ "model.language_model.layers.46.mlp_hyper_connection.input_mix_weight_up",
875
+ "model.language_model.layers.47.attn_hyper_connection.block_inject_weight",
876
+ "model.language_model.layers.47.attn_hyper_connection.input_mix_weight_down",
877
+ "model.language_model.layers.47.attn_hyper_connection.input_mix_weight_up",
878
+ "model.language_model.layers.47.mlp.gate",
879
+ "model.language_model.layers.47.mlp.shared_expert.down_proj",
880
+ "model.language_model.layers.47.mlp.shared_expert.gate_proj",
881
+ "model.language_model.layers.47.mlp.shared_expert.up_proj",
882
+ "model.language_model.layers.47.mlp.shared_expert_gate",
883
+ "model.language_model.layers.47.mlp_hyper_connection.block_inject_weight",
884
+ "model.language_model.layers.47.mlp_hyper_connection.input_mix_weight_down",
885
+ "model.language_model.layers.47.mlp_hyper_connection.input_mix_weight_up",
886
+ "model.language_model.layers.47.self_attn.indexer.index_qk_proj",
887
+ "model.language_model.layers.47.self_attn.k_proj",
888
+ "model.language_model.layers.47.self_attn.o_proj",
889
+ "model.language_model.layers.47.self_attn.q_proj",
890
+ "model.language_model.layers.47.self_attn.v_proj",
891
+ "model.language_model.layers.5.attn_hyper_connection.block_inject_weight",
892
+ "model.language_model.layers.5.attn_hyper_connection.input_mix_weight_down",
893
+ "model.language_model.layers.5.attn_hyper_connection.input_mix_weight_up",
894
+ "model.language_model.layers.5.linear_attn.conv1d",
895
+ "model.language_model.layers.5.linear_attn.in_proj_a",
896
+ "model.language_model.layers.5.linear_attn.in_proj_b",
897
+ "model.language_model.layers.5.linear_attn.in_proj_qkv",
898
+ "model.language_model.layers.5.linear_attn.in_proj_z",
899
+ "model.language_model.layers.5.linear_attn.out_proj",
900
+ "model.language_model.layers.5.mlp.gate",
901
+ "model.language_model.layers.5.mlp.shared_expert.down_proj",
902
+ "model.language_model.layers.5.mlp.shared_expert.gate_proj",
903
+ "model.language_model.layers.5.mlp.shared_expert.up_proj",
904
+ "model.language_model.layers.5.mlp.shared_expert_gate",
905
+ "model.language_model.layers.5.mlp_hyper_connection.block_inject_weight",
906
+ "model.language_model.layers.5.mlp_hyper_connection.input_mix_weight_down",
907
+ "model.language_model.layers.5.mlp_hyper_connection.input_mix_weight_up",
908
+ "model.language_model.layers.6.attn_hyper_connection.block_inject_weight",
909
+ "model.language_model.layers.6.attn_hyper_connection.input_mix_weight_down",
910
+ "model.language_model.layers.6.attn_hyper_connection.input_mix_weight_up",
911
+ "model.language_model.layers.6.linear_attn.conv1d",
912
+ "model.language_model.layers.6.linear_attn.in_proj_a",
913
+ "model.language_model.layers.6.linear_attn.in_proj_b",
914
+ "model.language_model.layers.6.linear_attn.in_proj_qkv",
915
+ "model.language_model.layers.6.linear_attn.in_proj_z",
916
+ "model.language_model.layers.6.linear_attn.out_proj",
917
+ "model.language_model.layers.6.mlp.gate",
918
+ "model.language_model.layers.6.mlp.shared_expert.down_proj",
919
+ "model.language_model.layers.6.mlp.shared_expert.gate_proj",
920
+ "model.language_model.layers.6.mlp.shared_expert.up_proj",
921
+ "model.language_model.layers.6.mlp.shared_expert_gate",
922
+ "model.language_model.layers.6.mlp_hyper_connection.block_inject_weight",
923
+ "model.language_model.layers.6.mlp_hyper_connection.input_mix_weight_down",
924
+ "model.language_model.layers.6.mlp_hyper_connection.input_mix_weight_up",
925
+ "model.language_model.layers.7.attn_hyper_connection.block_inject_weight",
926
+ "model.language_model.layers.7.attn_hyper_connection.input_mix_weight_down",
927
+ "model.language_model.layers.7.attn_hyper_connection.input_mix_weight_up",
928
+ "model.language_model.layers.7.mlp.gate",
929
+ "model.language_model.layers.7.mlp.shared_expert.down_proj",
930
+ "model.language_model.layers.7.mlp.shared_expert.gate_proj",
931
+ "model.language_model.layers.7.mlp.shared_expert.up_proj",
932
+ "model.language_model.layers.7.mlp.shared_expert_gate",
933
+ "model.language_model.layers.7.mlp_hyper_connection.block_inject_weight",
934
+ "model.language_model.layers.7.mlp_hyper_connection.input_mix_weight_down",
935
+ "model.language_model.layers.7.mlp_hyper_connection.input_mix_weight_up",
936
+ "model.language_model.layers.7.self_attn.indexer.index_qk_proj",
937
+ "model.language_model.layers.7.self_attn.k_proj",
938
+ "model.language_model.layers.7.self_attn.o_proj",
939
+ "model.language_model.layers.7.self_attn.q_proj",
940
+ "model.language_model.layers.7.self_attn.v_proj",
941
+ "model.language_model.layers.8.attn_hyper_connection.block_inject_weight",
942
+ "model.language_model.layers.8.attn_hyper_connection.input_mix_weight_down",
943
+ "model.language_model.layers.8.attn_hyper_connection.input_mix_weight_up",
944
+ "model.language_model.layers.8.linear_attn.conv1d",
945
+ "model.language_model.layers.8.linear_attn.in_proj_a",
946
+ "model.language_model.layers.8.linear_attn.in_proj_b",
947
+ "model.language_model.layers.8.linear_attn.in_proj_qkv",
948
+ "model.language_model.layers.8.linear_attn.in_proj_z",
949
+ "model.language_model.layers.8.linear_attn.out_proj",
950
+ "model.language_model.layers.8.mlp.gate",
951
+ "model.language_model.layers.8.mlp.shared_expert.down_proj",
952
+ "model.language_model.layers.8.mlp.shared_expert.gate_proj",
953
+ "model.language_model.layers.8.mlp.shared_expert.up_proj",
954
+ "model.language_model.layers.8.mlp.shared_expert_gate",
955
+ "model.language_model.layers.8.mlp_hyper_connection.block_inject_weight",
956
+ "model.language_model.layers.8.mlp_hyper_connection.input_mix_weight_down",
957
+ "model.language_model.layers.8.mlp_hyper_connection.input_mix_weight_up",
958
+ "model.language_model.layers.9.attn_hyper_connection.block_inject_weight",
959
+ "model.language_model.layers.9.attn_hyper_connection.input_mix_weight_down",
960
+ "model.language_model.layers.9.attn_hyper_connection.input_mix_weight_up",
961
+ "model.language_model.layers.9.linear_attn.conv1d",
962
+ "model.language_model.layers.9.linear_attn.in_proj_a",
963
+ "model.language_model.layers.9.linear_attn.in_proj_b",
964
+ "model.language_model.layers.9.linear_attn.in_proj_qkv",
965
+ "model.language_model.layers.9.linear_attn.in_proj_z",
966
+ "model.language_model.layers.9.linear_attn.out_proj",
967
+ "model.language_model.layers.9.mlp.gate",
968
+ "model.language_model.layers.9.mlp.shared_expert.down_proj",
969
+ "model.language_model.layers.9.mlp.shared_expert.gate_proj",
970
+ "model.language_model.layers.9.mlp.shared_expert.up_proj",
971
+ "model.language_model.layers.9.mlp.shared_expert_gate",
972
+ "model.language_model.layers.9.mlp_hyper_connection.block_inject_weight",
973
+ "model.language_model.layers.9.mlp_hyper_connection.input_mix_weight_down",
974
+ "model.language_model.layers.9.mlp_hyper_connection.input_mix_weight_up",
975
+ "model.visual.blocks.0.attn.proj",
976
+ "model.visual.blocks.0.attn.qkv",
977
+ "model.visual.blocks.0.mlp.linear_fc1",
978
+ "model.visual.blocks.0.mlp.linear_fc2",
979
+ "model.visual.blocks.1.attn.proj",
980
+ "model.visual.blocks.1.attn.qkv",
981
+ "model.visual.blocks.1.mlp.linear_fc1",
982
+ "model.visual.blocks.1.mlp.linear_fc2",
983
+ "model.visual.blocks.10.attn.proj",
984
+ "model.visual.blocks.10.attn.qkv",
985
+ "model.visual.blocks.10.mlp.linear_fc1",
986
+ "model.visual.blocks.10.mlp.linear_fc2",
987
+ "model.visual.blocks.11.attn.proj",
988
+ "model.visual.blocks.11.attn.qkv",
989
+ "model.visual.blocks.11.mlp.linear_fc1",
990
+ "model.visual.blocks.11.mlp.linear_fc2",
991
+ "model.visual.blocks.12.attn.proj",
992
+ "model.visual.blocks.12.attn.qkv",
993
+ "model.visual.blocks.12.mlp.linear_fc1",
994
+ "model.visual.blocks.12.mlp.linear_fc2",
995
+ "model.visual.blocks.13.attn.proj",
996
+ "model.visual.blocks.13.attn.qkv",
997
+ "model.visual.blocks.13.mlp.linear_fc1",
998
+ "model.visual.blocks.13.mlp.linear_fc2",
999
+ "model.visual.blocks.14.attn.proj",
1000
+ "model.visual.blocks.14.attn.qkv",
1001
+ "model.visual.blocks.14.mlp.linear_fc1",
1002
+ "model.visual.blocks.14.mlp.linear_fc2",
1003
+ "model.visual.blocks.15.attn.proj",
1004
+ "model.visual.blocks.15.attn.qkv",
1005
+ "model.visual.blocks.15.mlp.linear_fc1",
1006
+ "model.visual.blocks.15.mlp.linear_fc2",
1007
+ "model.visual.blocks.16.attn.proj",
1008
+ "model.visual.blocks.16.attn.qkv",
1009
+ "model.visual.blocks.16.mlp.linear_fc1",
1010
+ "model.visual.blocks.16.mlp.linear_fc2",
1011
+ "model.visual.blocks.17.attn.proj",
1012
+ "model.visual.blocks.17.attn.qkv",
1013
+ "model.visual.blocks.17.mlp.linear_fc1",
1014
+ "model.visual.blocks.17.mlp.linear_fc2",
1015
+ "model.visual.blocks.18.attn.proj",
1016
+ "model.visual.blocks.18.attn.qkv",
1017
+ "model.visual.blocks.18.mlp.linear_fc1",
1018
+ "model.visual.blocks.18.mlp.linear_fc2",
1019
+ "model.visual.blocks.19.attn.proj",
1020
+ "model.visual.blocks.19.attn.qkv",
1021
+ "model.visual.blocks.19.mlp.linear_fc1",
1022
+ "model.visual.blocks.19.mlp.linear_fc2",
1023
+ "model.visual.blocks.2.attn.proj",
1024
+ "model.visual.blocks.2.attn.qkv",
1025
+ "model.visual.blocks.2.mlp.linear_fc1",
1026
+ "model.visual.blocks.2.mlp.linear_fc2",
1027
+ "model.visual.blocks.20.attn.proj",
1028
+ "model.visual.blocks.20.attn.qkv",
1029
+ "model.visual.blocks.20.mlp.linear_fc1",
1030
+ "model.visual.blocks.20.mlp.linear_fc2",
1031
+ "model.visual.blocks.21.attn.proj",
1032
+ "model.visual.blocks.21.attn.qkv",
1033
+ "model.visual.blocks.21.mlp.linear_fc1",
1034
+ "model.visual.blocks.21.mlp.linear_fc2",
1035
+ "model.visual.blocks.22.attn.proj",
1036
+ "model.visual.blocks.22.attn.qkv",
1037
+ "model.visual.blocks.22.mlp.linear_fc1",
1038
+ "model.visual.blocks.22.mlp.linear_fc2",
1039
+ "model.visual.blocks.23.attn.proj",
1040
+ "model.visual.blocks.23.attn.qkv",
1041
+ "model.visual.blocks.23.mlp.linear_fc1",
1042
+ "model.visual.blocks.23.mlp.linear_fc2",
1043
+ "model.visual.blocks.24.attn.proj",
1044
+ "model.visual.blocks.24.attn.qkv",
1045
+ "model.visual.blocks.24.mlp.linear_fc1",
1046
+ "model.visual.blocks.24.mlp.linear_fc2",
1047
+ "model.visual.blocks.25.attn.proj",
1048
+ "model.visual.blocks.25.attn.qkv",
1049
+ "model.visual.blocks.25.mlp.linear_fc1",
1050
+ "model.visual.blocks.25.mlp.linear_fc2",
1051
+ "model.visual.blocks.26.attn.proj",
1052
+ "model.visual.blocks.26.attn.qkv",
1053
+ "model.visual.blocks.26.mlp.linear_fc1",
1054
+ "model.visual.blocks.26.mlp.linear_fc2",
1055
+ "model.visual.blocks.3.attn.proj",
1056
+ "model.visual.blocks.3.attn.qkv",
1057
+ "model.visual.blocks.3.mlp.linear_fc1",
1058
+ "model.visual.blocks.3.mlp.linear_fc2",
1059
+ "model.visual.blocks.4.attn.proj",
1060
+ "model.visual.blocks.4.attn.qkv",
1061
+ "model.visual.blocks.4.mlp.linear_fc1",
1062
+ "model.visual.blocks.4.mlp.linear_fc2",
1063
+ "model.visual.blocks.5.attn.proj",
1064
+ "model.visual.blocks.5.attn.qkv",
1065
+ "model.visual.blocks.5.mlp.linear_fc1",
1066
+ "model.visual.blocks.5.mlp.linear_fc2",
1067
+ "model.visual.blocks.6.attn.proj",
1068
+ "model.visual.blocks.6.attn.qkv",
1069
+ "model.visual.blocks.6.mlp.linear_fc1",
1070
+ "model.visual.blocks.6.mlp.linear_fc2",
1071
+ "model.visual.blocks.7.attn.proj",
1072
+ "model.visual.blocks.7.attn.qkv",
1073
+ "model.visual.blocks.7.mlp.linear_fc1",
1074
+ "model.visual.blocks.7.mlp.linear_fc2",
1075
+ "model.visual.blocks.8.attn.proj",
1076
+ "model.visual.blocks.8.attn.qkv",
1077
+ "model.visual.blocks.8.mlp.linear_fc1",
1078
+ "model.visual.blocks.8.mlp.linear_fc2",
1079
+ "model.visual.blocks.9.attn.proj",
1080
+ "model.visual.blocks.9.attn.qkv",
1081
+ "model.visual.blocks.9.mlp.linear_fc1",
1082
+ "model.visual.blocks.9.mlp.linear_fc2",
1083
+ "model.visual.merger.linear_fc1",
1084
+ "model.visual.merger.linear_fc2",
1085
+ "model.visual.patch_embed.proj",
1086
+ "model.visual.pos_embed",
1087
+ "mtp.fc_embedding",
1088
+ "mtp.fc_hidden",
1089
+ "mtp.hyper_connection_mixer.input_mix_weight_down",
1090
+ "mtp.hyper_connection_mixer.input_mix_weight_up",
1091
+ "mtp.layers.0.attn_hyper_connection.block_inject_weight",
1092
+ "mtp.layers.0.attn_hyper_connection.input_mix_weight_down",
1093
+ "mtp.layers.0.attn_hyper_connection.input_mix_weight_up",
1094
+ "mtp.layers.0.mlp.gate",
1095
+ "mtp.layers.0.mlp.shared_expert.down_proj",
1096
+ "mtp.layers.0.mlp.shared_expert.gate_proj",
1097
+ "mtp.layers.0.mlp.shared_expert.up_proj",
1098
+ "mtp.layers.0.mlp.shared_expert_gate",
1099
+ "mtp.layers.0.mlp_hyper_connection.block_inject_weight",
1100
+ "mtp.layers.0.mlp_hyper_connection.input_mix_weight_down",
1101
+ "mtp.layers.0.mlp_hyper_connection.input_mix_weight_up",
1102
+ "mtp.layers.0.self_attn.indexer.index_qk_proj",
1103
+ "mtp.layers.0.self_attn.k_proj",
1104
+ "mtp.layers.0.self_attn.o_proj",
1105
+ "mtp.layers.0.self_attn.q_proj",
1106
+ "mtp.layers.0.self_attn.v_proj"
1107
+ ]
1108
+ }
1109
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 248044,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 248046,
6
+ 248044
7
+ ],
8
+ "pad_token_id": 248044,
9
+ "temperature": 0.6,
10
+ "top_k": 20,
11
+ "top_p": 0.95
12
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:774f0ceeadb40d165f2b3ff397d5f3840e6ca8fcb8f3d39d8acb4fea9e52c941
3
+ size 1040155912
model-00002-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6841fe21fa8a8a7a693c585efe65cd2732889095b696da88bda0cb287366910b
3
+ size 1678209208
model-00003-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:974a2a2ab551f8f1405a4955ab32a8721c68c73dd85b382491d9f0e6a34ee752
3
+ size 993901136
model-00004-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:591f488ab5cd5f0bd4fe28266099523761b9b9339137800734c34ffd84595538
3
+ size 1678209208
model-00005-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c50bf465a4a0129f1a1196c0e75d881048238f7107b356fc734007eac0d3b123
3
+ size 1717267690
model-00006-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57f103367fe36c9dc355c648d521d026caabb4ce5ba51038c78bd935f6593736
3
+ size 1600008328
model-00007-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cde2a12854770c706a74a5c7b85ffc140e3d4dba93dd4bee1cd7aa07cc348c90
3
+ size 1600008328
model-00008-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98ae323b5a61abdcf58a0dc9db7eb1a5a748cebc1ef1932804a1411500c09d84
3
+ size 1600008328
model-00009-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70f4b66bca5c4c0b441cf4284d4ab311670e182bfc5ec51c7294a4054baeb20f
3
+ size 1600008328
model-00010-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d5dfa1dfa9a6c482f100f6e75c1f5e8ac02391c62e844e53f4a53996baf2206
3
+ size 1600008328
model-00011-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de085adb563e530e7ad157220cd1a219be189a1db0fcc3f2507886a2d78bd4e1
3
+ size 1600008328
model-00012-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97dac7c366ba19858d78c01364054136449ccb3cf60768587461cbf0bb582c90
3
+ size 1600008328
model-00013-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:467be10b473d25e0581f8fb1bc986916c8b9301272f72e7a05762484ffc080a9
3
+ size 1600008328
model-00014-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38def6e9bbe6f242c2f92858c3e9a2b356d74d0180e323677158016dbee32ae7
3
+ size 1600008328
model-00015-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90c6c665f1f819bc6350714652c8fb77064fec57832ef51b878a0cdf9f602236
3
+ size 1600008328
model-00016-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88185c2530641b37dcad80d37b505c5257035e87ad29fa86640aae4814903e4f
3
+ size 1600008328
model-00017-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42e3b96359116d7471a74cb7de08f6a4e9e173830d562f1f74ff09d1d253ad38
3
+ size 1600008328
model-00018-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:040adc7270efca6498c6f6a07dd7da14925d20217fe6de67223d6856fb4f2ad7
3
+ size 1600008328
model-00019-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8a831992deb1a694439e37b7136ff2e181457a8683bff65f1f4b99772cabe58
3
+ size 1600008328
model-00020-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9426112fbd49ef4f5e9dd38dc2f93b046602e6d92cff732d33b96413ff1b40dc
3
+ size 1600008328
model-00021-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fce8426665f749ef38c4fad33b612f892b728c0f74fe60cab0e43ece785b4e77
3
+ size 1600008328
model-00022-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a20dd415f280774321538e6b9e78fba9f8e218a9090eae52ea7172b85b9c5a6d
3
+ size 1600008328
model-00023-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4100591839471a8721d9cedb251554dcef5ebc1dfd96ea7dce8912dd1e9921b2
3
+ size 1600008328
model-00024-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbd6d7a4b54b6a6032e15b7c2fc9d6526377f89992d61da8837ad059f3210b45
3
+ size 1600008328
model-00025-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9fa7b04de155dd2be74b5693d5b36e5463dabf956b9be36f67d4aae6b8dd304
3
+ size 1600008328
model-00026-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:193c6b9da2c6b69c34d8a14b65234871db272761e712c2e56bac3f13c2d0cbf7
3
+ size 1600008328
model-00027-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7efa372552c07ee8783e27c4b8ebd12fc75333f92553861bec8c7c0155a03fe
3
+ size 1600008328
model-00028-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fc13d084d62c06b928dfb388a1b36f5c27cd25bd3cbb68a0b21e1c4cfc68e6e
3
+ size 1600008328
model-00029-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a11f5a6360b349159575871db5c553c40c37be68382b82551a2c2372f15f2811
3
+ size 1600008328
model-00030-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cac780a675f87fc699435a5207cbecb30a887271b5514b60b7e45d9bebe0e1ec
3
+ size 1600008336
model-00031-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d390a872eb3928b9d7c0a720ced792cc111613812a163d9fd6c97fd35d633897
3
+ size 1600008336
model-00032-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b36fe4e7215a12cdc6c7499aafde51cfbce51266785184b99d3838a6b37b2c8
3
+ size 1600008336
model-00033-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fb0504f406c0c71ee40e9e158492973c5c35337889c81b974892b407620c64c
3
+ size 1600008336
model-00034-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d989a812a8d8d9330c8c4d21d7be98a0114705ca3bd7834d43bb268bad6be08
3
+ size 1600008336
model-00035-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e56ba0a714198bd07a2d8fe9d91fbf5c72601c1b98edf09b289a2538445cb75
3
+ size 1600008336
model-00036-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c23fd6f6cfa9acfa961ba26e7c9c0dd7b05782c5bbf8f55e2526119b5c516fd4
3
+ size 1600008336
model-00037-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b824a11bf460bd78c068e3d301e914a0bec2f243a6eea4679be7200d179248e7
3
+ size 942343240
model-00038-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:716eda81a763ffdefd1ac5306b4c80eef123c4dee24ed3a565a22438497581cb
3
+ size 1678211256
model-00039-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b91e3bfbcd765ce704db85b78f69055da59456ae733df635e3973ff2f966a148
3
+ size 877983696
model-00040-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3311ed7821b68bddb1542f06ca349feec72e68288113fe4899b08a3b8156343f
3
+ size 1678211256
model-00041-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d79949b61730a0fa1f6cd1af149fa50fedc051566cdea48283de8c7a32d972c
3
+ size 1103350296
model-00042-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f61684338e203cb2ccff6e451eda5705e10043d386b7ed55d4732d23eb5c77b
3
+ size 1678211256
model-00043-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bab92cfefbb811ee63c4e32595024117c7df3537f5359606eb1de23fe4f115b2
3
+ size 1000455912
model-00044-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d05f138848edca4138c3d1c6be4a432a0817eb29fa65f12d8611ecc7685c493
3
+ size 1678211256
model-00045-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4610384e28f7a91b8bf2a1527178c82cbe7d50ecd9a33b61ee878fa8e8bced85
3
+ size 993902176
model-00046-of-00131.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ee9b0e45810300cbd4fd6f2acfe85080e7032a9ce5d77038b86f357b6700c23
3
+ size 1678211256