{ "activation_function": "gelu_new", "architectures": [ "PhiForCausalLM" ], "attention_dropout": 0.0, "attn_pdrop": 0.0, "embd_pdrop": 0.0, "flash_attn": false, "flash_rotary": false, "fused_dense": false, "hidden_act": "gelu_new", "initializer_range": 0.02, "intermediate_size": 8192, "layer_norm_eps": 1e-05, "layer_norm_epsilon": 1e-05, "model_type": "phi-causal-lm", "n_embd": 1024, "n_head": 16, "n_head_kv": null, "n_inner": null, "n_layer": 1, "n_positions": 2048, "num_key_value_heads": null, "pad_vocab_size_multiple": 64, "partial_rotary_factor": 0.5, "qk_layernorm": false, "resid_pdrop": 0.0, "rope_scaling": null, "rope_theta": 10000.0, "rotary_dim": 32, "tie_word_embeddings": false, "torch_dtype": "float32", "transformers_version": "4.50.0", "use_cache": true, "vocab_size": 51200 }