SystemAdmin123/tiny-random-LlamaForCausalLM

1{2  "_name_or_path": "trl-internal-testing/tiny-random-LlamaForCausalLM",3  "architectures": [4    "LlamaForCausalLM"5  ],6  "attention_bias": false,7  "attention_dropout": 0.0,8  "bos_token_id": 0,9  "eos_token_id": 2,10  "head_dim": 4,11  "hidden_act": "silu",12  "hidden_size": 16,13  "initializer_range": 0.02,14  "intermediate_size": 64,15  "max_position_embeddings": 2048,16  "mlp_bias": false,17  "model_type": "llama",18  "num_attention_heads": 4,19  "num_hidden_layers": 2,20  "num_key_value_heads": 4,21  "pad_token_id": -1,22  "pretraining_tp": 1,23  "rms_norm_eps": 1e-06,24  "rope_scaling": null,25  "rope_theta": 10000.0,26  "tie_word_embeddings": false,27  "torch_dtype": "bfloat16",28  "transformers_version": "4.48.1",29  "use_cache": false,30  "vocab_size": 3200031}32