jradchenko/DeciCoder-1b
017
1{2 "architectures": [3 "DeciCoderForCausalLM"4 ],5 "auto_map": {6 "AutoConfig": "configuration_llama.LlamaConfig",7 "AutoModel": "modeling_llama.LlamaModel",8 "AutoModelForCausalLM": "modeling_llama.LlamaForCausalLM",9 "AutoModelForSequenceClassification": "modeling_llama.LlamaForSequenceClassification"10 },11 "bos_token_id": 0,12 "eos_token_id": 0,13 "hidden_act": "silu",14 "hidden_size": 2048,15 "initializer_range": 0.02,16 "intermediate_size": 5888,17 "max_position_embeddings": 2048,18 "num_attention_heads": 32,19 "num_hidden_layers": 20,20 "num_key_value_heads": 4,21 "pretraining_tp": 1,22 "rms_norm_eps": 1e-05,23 "rope_scaling": null,24 "tie_word_embeddings": false,25 "torch_dtype": "bfloat16",26 "use_bfloat16": true,27 "transformers_version": "4.31.0.dev0",28 "use_cache": true,29 "vocab_size": 4915230}31 