trl-internal-testing/tiny-GPT2LMHeadModel

1{2  "activation_function": "gelu_new",3  "architectures": [4    "GPT2LMHeadModel"5  ],6  "attn_pdrop": 0.1,7  "bos_token_id": 50256,8  "dtype": "float32",9  "embd_pdrop": 0.1,10  "eos_token_id": 50256,11  "initializer_range": 0.02,12  "intermediate_size": 32,13  "layer_norm_epsilon": 1e-05,14  "model_type": "gpt2",15  "n_embd": 8,16  "n_head": 4,17  "n_inner": null,18  "n_layer": 2,19  "n_positions": 1024,20  "num_key_value_heads": 2,21  "reorder_and_upcast_attn": false,22  "resid_pdrop": 0.1,23  "scale_attn_by_inverse_layer_idx": false,24  "scale_attn_weights": true,25  "summary_activation": null,26  "summary_first_dropout": 0.1,27  "summary_proj_to_labels": true,28  "summary_type": "cls_index",29  "summary_use_proj": true,30  "transformers_version": "4.57.3",31  "use_cache": true,32  "vocab_size": 5025733}34