|
{ |
|
"_name_or_path": "alycialee/m2-bert-341M", |
|
"alibi_starting_size": 512, |
|
"architectures": [ |
|
"BertForMaskedLM" |
|
], |
|
"attention_probs_dropout_prob": 0.0, |
|
"bidirectional": true, |
|
"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": 1792, |
|
"initializer_range": 0.02, |
|
"intermediate_size": 7168, |
|
"layer_norm_eps": 1e-12, |
|
"max_position_embeddings": 128, |
|
"model_type": "bert", |
|
"num_attention_heads": 12, |
|
"num_hidden_layers": 12, |
|
"pad_token_id": 0, |
|
"position_embedding_type": "absolute", |
|
"torch_dtype": "float32", |
|
"transformers_version": "4.28.1", |
|
"type_vocab_size": 2, |
|
"use_cache": true, |
|
"long_conv_l_max": 128, |
|
"long_conv_kernel_learning_rate": 1e-3, |
|
"hyena_lr_pos_emb": 1e-5, |
|
"hyena_w": 10, |
|
"hyena_wd": 0.1, |
|
"hyena_emb_dim": 5, |
|
"hyena_filter_order": 128, |
|
"residual_long_conv": true, |
|
"use_glu_mlp": true, |
|
"use_monarch_mlp": true, |
|
"monarch_mlp_nblocks": 4, |
|
"use_positional_encodings" : true, |
|
"vocab_size": 30528 |
|
} |