config.json

{
  "_name_or_path": "ayjays132/CustomGPT2Conversational",
  "activation_function": "gelu_new",
  "architectures": [
    "GPT2LMHeadModel"
  ],
  "attn_pdrop": 0.1,
  "bos_token_id": 50256,
  "config": {
    "activation_function": "gelu_new",
    "attn_pdrop": 0.1,
    "embd_pdrop": 0.1,
    "gradient_checkpointing": true,
    "initializer_range": 0.02,
    "layer_norm_epsilon": 1e-05,
    "n_ctx": 2048,
    "n_embd": 2048,
    "n_head": 16,
    "n_layer": 36,
    "n_positions": 2048,
    "resid_pdrop": 0.1,
    "scale_attn_weights": true,
    "use_cache": true,
    "vocab_size": 50257
  },
  "embd_pdrop": 0.1,
  "eos_token_id": 50256,
  "initializer_range": 0.02,
  "language": "en",
  "layer_norm_epsilon": 1e-05,
  "library_name": "transformers",
  "license": "apache-2.0",
  "metrics": [
    "perplexity",
    "accuracy"
  ],
  "model_type": "gpt2",
  "n_embd": 768,
  "n_head": 12,
  "n_inner": null,
  "n_layer": 12,
  "n_positions": 1024,
  "pipeline_tag": "conversational",
  "reorder_and_upcast_attn": false,
  "resid_pdrop": 0.1,
  "scale_attn_by_inverse_layer_idx": false,
  "scale_attn_weights": true,
  "summary_activation": null,
  "summary_first_dropout": 0.1,
  "summary_proj_to_labels": true,
  "summary_type": "cls_index",
  "summary_use_proj": true,
  "tags": [
    "conversational",
    "state-of-the-art"
  ],
  "task_specific_params": {
    "conversational": {
      "do_sample": true,
      "early_stopping": true,
      "frequency_penalty": 0.5,
      "length_penalty": 2.0,
      "max_length": 1024,
      "min_length": 20,
      "no_repeat_ngram_size": 3,
      "num_beams": 5,
      "presence_penalty": 0.5,
      "temperature": 0.7,
      "top_k": 40,
      "top_p": 0.95
    }
  },
  "tokenizer_config": {
    "bos_token_id": 50256,
    "eos_token_id": 50256,
    "n_positions": 2048,
    "padding_side": "left",
    "truncation_side": "right"
  },
  "torch_dtype": "float32",
  "transformers_version": "4.37.2",
  "use_cache": true,
  "vocab_size": 50257
}