| { | |
| "_name_or_path": "Tweeties/tweety-tatar-hydra-base-7b-2024-v1", | |
| "additional_special_tokens_ids": [], | |
| "architectures": [ | |
| "LlamaHydraForCausalLM" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "modeling_llama_hydra.LlamaHydraConfig", | |
| "AutoModelForCausalLM": "modeling_llama_hydra.LlamaHydraForCausalLM" | |
| }, | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 1, | |
| "cls_token_id": null, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 4096, | |
| "initializer_range": 0.02, | |
| "input_vocab_size": 64007, | |
| "intermediate_size": 11008, | |
| "mask_token_id": null, | |
| "max_position_embeddings": 4096, | |
| "model_type": "llama_hydra", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 32, | |
| "output_vocab_size": 32000, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "tokenizer_class": "LlamaTokenizerFast", | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.37.1", | |
| "unk_token_id": 0, | |
| "use_cache": true, | |
| "vocab_size": 32000 | |
| } | |