CodeAtCMU/SmolLM2-360M_full_sft_code_data_120K_remove_comments
015
1{2 "architectures": [3 "LlamaForCausalLM"4 ],5 "attention_bias": false,6 "attention_dropout": 0.0,7 "bos_token_id": 0,8 "eos_token_id": 0,9 "head_dim": 64,10 "hidden_act": "silu",11 "hidden_size": 960,12 "initializer_range": 0.02,13 "intermediate_size": 2560,14 "is_llama_config": true,15 "max_position_embeddings": 8192,16 "mlp_bias": false,17 "model_type": "llama",18 "num_attention_heads": 15,19 "num_hidden_layers": 32,20 "num_key_value_heads": 5,21 "pretraining_tp": 1,22 "rms_norm_eps": 1e-05,23 "rope_interleaved": false,24 "rope_scaling": null,25 "rope_theta": 100000,26 "tie_word_embeddings": true,27 "torch_dtype": "bfloat16",28 "transformers_version": "4.51.3",29 "use_cache": false,30 "vocab_size": 4915231}32 