trl-internal-testing/tiny-LlamaForSequenceClassification-3.2

1{2  "architectures": [3    "LlamaForSequenceClassification"4  ],5  "attention_bias": false,6  "attention_dropout": 0.0,7  "bos_token_id": 128000,8  "dtype": "bfloat16",9  "eos_token_id": [10    128001,11    128008,12    12800913  ],14  "head_dim": 64,15  "hidden_act": "silu",16  "hidden_size": 16,17  "id2label": {18    "0": "LABEL_0"19  },20  "initializer_range": 0.02,21  "intermediate_size": 32,22  "label2id": {23    "LABEL_0": 024  },25  "max_position_embeddings": 131072,26  "mlp_bias": false,27  "model_type": "llama",28  "num_attention_heads": 4,29  "num_hidden_layers": 2,30  "num_key_value_heads": 2,31  "pretraining_tp": 1,32  "rms_norm_eps": 1e-05,33  "rope_scaling": {34    "factor": 32.0,35    "high_freq_factor": 4.0,36    "low_freq_factor": 1.0,37    "original_max_position_embeddings": 8192,38    "rope_type": "llama3"39  },40  "rope_theta": 500000.0,41  "tie_word_embeddings": true,42  "transformers_version": "4.57.3",43  "use_cache": true,44  "vocab_size": 12825645}46