entropy/roberta_zinc_decoder
1103
1{2 "_name_or_path": "mean_roberta/checkpoint-50000/",3 "activation_function": "gelu_new",4 "add_cross_attention": true,5 "architectures": [6 "ConditionalGPT2LMHeadModel"7 ],8 "attn_pdrop": 0.1,9 "auto_map": {10 "AutoModelForCausalLM": "conditional_gpt2_model.ConditionalGPT2LMHeadModel"11 },12 "bos_token_id": 0,13 "embd_pdrop": 0.1,14 "eos_token_id": 2,15 "initializer_range": 0.02,16 "layer_norm_epsilon": 1e-05,17 "model_type": "gpt2",18 "n_embd": 768,19 "n_head": 8,20 "n_inner": null,21 "n_layer": 6,22 "n_positions": 256,23 "reorder_and_upcast_attn": false,24 "resid_pdrop": 0.1,25 "scale_attn_by_inverse_layer_idx": false,26 "scale_attn_weights": true,27 "summary_activation": null,28 "summary_first_dropout": 0.1,29 "summary_proj_to_labels": true,30 "summary_type": "cls_index",31 "summary_use_proj": true,32 "torch_dtype": "float32",33 "transformers_version": "4.28.1",34 "use_cache": true,35 "vocab_size": 270736}37 