| { | |
| "architectures": [ | |
| "LlamaForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_darwinlm.DarwinLMConfig", | |
| "AutoModelForCausalLM": "modeling_darwinlm.LlamaForCausalLM" | |
| }, | |
| "bos_token_id": 1, | |
| "dim_each_mlp": { | |
| "0.mlp.down_proj": 3104, | |
| "1.mlp.down_proj": 8032, | |
| "10.mlp.down_proj": 1824, | |
| "11.mlp.down_proj": 0, | |
| "12.mlp.down_proj": 3104, | |
| "13.mlp.down_proj": 5280, | |
| "14.mlp.down_proj": 5280, | |
| "15.mlp.down_proj": 4256, | |
| "16.mlp.down_proj": 4256, | |
| "17.mlp.down_proj": 6496, | |
| "18.mlp.down_proj": 6496, | |
| "19.mlp.down_proj": 5280, | |
| "2.mlp.down_proj": 6496, | |
| "20.mlp.down_proj": 3104, | |
| "21.mlp.down_proj": 4256, | |
| "22.mlp.down_proj": 4256, | |
| "23.mlp.down_proj": 3104, | |
| "24.mlp.down_proj": 3104, | |
| "25.mlp.down_proj": 3104, | |
| "26.mlp.down_proj": 4256, | |
| "27.mlp.down_proj": 3104, | |
| "28.mlp.down_proj": 3104, | |
| "29.mlp.down_proj": 3104, | |
| "3.mlp.down_proj": 4256, | |
| "30.mlp.down_proj": 6496, | |
| "31.mlp.down_proj": 6496, | |
| "4.mlp.down_proj": 5280, | |
| "5.mlp.down_proj": 5280, | |
| "6.mlp.down_proj": 4256, | |
| "7.mlp.down_proj": 3104, | |
| "8.mlp.down_proj": 5280, | |
| "9.mlp.down_proj": 4256 | |
| }, | |
| "eos_token_id": 2, | |
| "head_dim": 128, | |
| "heads_each_attn": { | |
| "0.self_attn.o_proj": 7, | |
| "1.self_attn.o_proj": 11, | |
| "10.self_attn.o_proj": 14, | |
| "11.self_attn.o_proj": 32, | |
| "12.self_attn.o_proj": 32, | |
| "13.self_attn.o_proj": 4, | |
| "14.self_attn.o_proj": 11, | |
| "15.self_attn.o_proj": 22, | |
| "16.self_attn.o_proj": 14, | |
| "17.self_attn.o_proj": 18, | |
| "18.self_attn.o_proj": 22, | |
| "19.self_attn.o_proj": 14, | |
| "2.self_attn.o_proj": 14, | |
| "20.self_attn.o_proj": 22, | |
| "21.self_attn.o_proj": 22, | |
| "22.self_attn.o_proj": 22, | |
| "23.self_attn.o_proj": 18, | |
| "24.self_attn.o_proj": 32, | |
| "25.self_attn.o_proj": 32, | |
| "26.self_attn.o_proj": 18, | |
| "27.self_attn.o_proj": 31, | |
| "28.self_attn.o_proj": 28, | |
| "29.self_attn.o_proj": 32, | |
| "3.self_attn.o_proj": 14, | |
| "30.self_attn.o_proj": 26, | |
| "31.self_attn.o_proj": 32, | |
| "4.self_attn.o_proj": 18, | |
| "5.self_attn.o_proj": 22, | |
| "6.self_attn.o_proj": 18, | |
| "7.self_attn.o_proj": 22, | |
| "8.self_attn.o_proj": 14, | |
| "9.self_attn.o_proj": 18 | |
| }, | |
| "hidden_act": "silu", | |
| "hidden_size": 4096, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 11008, | |
| "kv_ignore": false, | |
| "max_position_embeddings": 2048, | |
| "mlp_bias": false, | |
| "model_type": "darwinlm", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 32, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.45.0.dev0", | |
| "use_cache": true, | |
| "vocab_size": 32000 | |
| } | |