bigcode/deepseekcoder-33b-codeqwen-align-subset
1194
1{2 "_name_or_path": "ds33b_self_align_codeqwen",3 "architectures": [4 "LlamaForCausalLM"5 ],6 "attention_bias": false,7 "attention_dropout": 0.0,8 "bos_token_id": 32013,9 "eos_token_id": 32014,10 "hidden_act": "silu",11 "hidden_size": 7168,12 "initializer_range": 0.02,13 "intermediate_size": 19200,14 "max_position_embeddings": 16384,15 "model_type": "llama",16 "num_attention_heads": 56,17 "num_hidden_layers": 62,18 "num_key_value_heads": 8,19 "pretraining_tp": 1,20 "rms_norm_eps": 1e-06,21 "rope_scaling": {22 "factor": 4.0,23 "type": "linear"24 },25 "rope_theta": 100000,26 "tie_word_embeddings": false,27 "torch_dtype": "float32",28 "transformers_version": "4.40.1",29 "use_cache": true,30 "vocab_size": 3225631}32 