| { | |
| "activation_function": "swiglu", | |
| "add_bias": true, | |
| "apply_residual_connection_post_layernorm": false, | |
| "architectures": [ | |
| "GraniteForCausalLM" | |
| ], | |
| "attention_head_type": "mha", | |
| "attention_multiplier": null, | |
| "attention_softmax_in_fp32": true, | |
| "attn_pdrop": 0.1, | |
| "auto_map": { | |
| "AutoConfig": "configuration_granite.GraniteConfig", | |
| "AutoModel": "modeling_granite.GraniteModel", | |
| "AutoModelForCausalLM": "modeling_granite.GraniteForCausalLM" | |
| }, | |
| "bos_token_id": 0, | |
| "embd_pdrop": 0.1, | |
| "eos_token_id": 0, | |
| "initializer_range": 0.02, | |
| "layer_norm_epsilon": 1e-05, | |
| "model_type": "granite", | |
| "n_embd": 2560, | |
| "n_head": 32, | |
| "n_inner": 10240, | |
| "n_layer": 32, | |
| "n_positions": 2048, | |
| "normalization_function": "rmsnorm", | |
| "num_key_value_heads": 32, | |
| "pad_token_id": 0, | |
| "position_embedding_type": "rope", | |
| "resid_pdrop": 0.1, | |
| "rope_theta": 10000, | |
| "scale_attention_softmax_in_fp32": true, | |
| "scale_attn_weights": true, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.38.1", | |
| "use_cache": true, | |
| "vocab_size": 49152 | |
| } | |