{
  "_name_or_path": "/network/scratch/z/zhixuan.lin/linear-rnn-torch/release/transformer-pro-760m-longcrawl64-48b",
  "architectures": [
    "ForgettingTransformerForCausalLM"
  ],
  "attention_bias": false,
  "bos_token_id": null,
  "decay_time_max": null,
  "decay_time_min": null,
  "elementwise_affine": true,
  "eos_token_id": null,
  "fgate_bias_init": false,
  "fgate_type": "none",
  "fuse_cross_entropy": true,
  "fuse_norm": true,
  "hidden_act": "swish",
  "hidden_ratio": 3.5,
  "hidden_size": 1536,
  "initializer_range": 0.02,
  "intermediate_size": null,
  "max_position_embeddings": null,
  "model_type": "forgetting_transformer-project_fox",
  "norm_eps": 1e-06,
  "num_heads": 12,
  "num_hidden_layers": 24,
  "num_kv_heads": null,
  "ogate_act": "sigmoid",
  "qk_norm": true,
  "qk_norm_share_param_across_head": true,
  "rope_base": 500000.0,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.44.0",
  "use_cache": true,
  "use_k_shift": true,
  "use_output_gate": true,
  "use_output_norm": true,
  "use_rope": true,
  "use_v_shift": true,
  "vocab_size": 50257,
  "window_size": null
}