Change attention_probs_dropout_prob to 0.1 so that FlashAttention/triton dependencies are avoided
ed2a544 | { | |
| "_name_or_path": "mosaicml/mosaic-bert", | |
| "alibi_starting_size": 512, | |
| "architectures": [ | |
| "BertForMaskedLM" | |
| ], | |
| "attention_probs_dropout_prob": 0.1, | |
| "auto_map": { | |
| "AutoConfig": "configuration_bert.BertConfig", | |
| "AutoModelForMaskedLM": "bert_layers.BertForMaskedLM" | |
| }, | |
| "classifier_dropout": null, | |
| "gradient_checkpointing": false, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 768, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 3072, | |
| "layer_norm_eps": 1e-12, | |
| "max_position_embeddings": 512, | |
| "model_type": "bert", | |
| "num_attention_heads": 12, | |
| "num_hidden_layers": 12, | |
| "pad_token_id": 0, | |
| "position_embedding_type": "absolute", | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.26.0", | |
| "type_vocab_size": 2, | |
| "use_cache": true, | |
| "vocab_size": 30528 | |
| } | |