{ "_name_or_path": "/data/scratch/haukur/surprisal/results/bytemistral.large.l16.d512.f2048.steps12k.bs256x1.lr0.0003.wd0.1", "architectures": [ "MistralForCausalLM" ], "attention_dropout": 0.1, "bos_token_id": 257, "eos_token_id": 258, "head_dim": 64, "hidden_act": "silu", "hidden_size": 512, "initializer_range": 0.02, "intermediate_size": 2048, "max_position_embeddings": 4096, "model_type": "mistral", "num_attention_heads": 8, "num_hidden_layers": 16, "num_key_value_heads": 1, "pad_token_id": 256, "rms_norm_eps": 1e-06, "rope_theta": 10000.0, "sliding_window": 4096, "tie_word_embeddings": false, "torch_dtype": "float32", "transformers_version": "4.47.1", "use_cache": true, "vocab_size": 260 }