| { | |
| "_name_or_path": "/home/c84379517/PLM-Team/plm-instruct-rl", | |
| "architectures": [ | |
| "EdgellmForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "attn_implementation": "flash_attention_2", | |
| "auto_map": { | |
| "AutoConfig": "configuration_edgellm.EdgellmConfig", | |
| "AutoModel": "modeling_edgellm.EdgellmForCausalLM", | |
| "AutoModelForCausalLM": "modeling_edgellm.EdgellmForCausalLM" | |
| }, | |
| "bos_token_id": 151643, | |
| "eos_token_id": 151643, | |
| "hidden_act": "relu2", | |
| "hidden_size": 2048, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 8192, | |
| "kv_lora_rank": 512, | |
| "max_position_embeddings": 4096, | |
| "model_type": "edgellm", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 32, | |
| "pretraining_tp": 1, | |
| "q_lora_rank": null, | |
| "qk_nope_head_dim": 128, | |
| "qk_rope_head_dim": 64, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 100000.0, | |
| "sliding_window": 4096, | |
| "tie_word_embeddings": true, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.46.1", | |
| "use_cache": false, | |
| "use_sliding_window": false, | |
| "v_head_dim": 128, | |
| "vocab_size": 151936 | |
| } | |