| { | |
| "_name_or_path": "/net/nfs.cirrascale/allennlp/claran/catwalk-models/branched-train-Business2018-2019-decon-110m", | |
| "activation_type": "swiglu", | |
| "alibi": false, | |
| "alibi_bias_max": 8.0, | |
| "architectures": [ | |
| "OLMoForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "attention_layer_norm": true, | |
| "attention_layer_norm_with_affine": true, | |
| "bias_for_layer_norm": null, | |
| "block_group_size": 1, | |
| "block_type": "sequential", | |
| "d_model": 768, | |
| "embedding_dropout": 0.0, | |
| "embedding_size": 50304, | |
| "eos_token_id": 50256, | |
| "flash_attention": false, | |
| "include_bias": true, | |
| "init_cutoff_factor": null, | |
| "init_device": null, | |
| "init_fn": "normal", | |
| "init_std": 0.02, | |
| "layer_norm_type": "low_precision", | |
| "layer_norm_with_affine": true, | |
| "max_sequence_length": 2048, | |
| "mlp_hidden_size": null, | |
| "mlp_ratio": 4, | |
| "model_type": "olmo", | |
| "multi_query_attention": false, | |
| "n_heads": 12, | |
| "n_layers": 12, | |
| "pad_token_id": 50256, | |
| "precision": null, | |
| "residual_dropout": 0.0, | |
| "rope": true, | |
| "rope_full_precision": true, | |
| "scale_logits": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.31.0", | |
| "use_cache": true, | |
| "vocab_size": 50280, | |
| "weight_tying": true | |
| } | |