CoolFace
Modelpublic

Salesforce/codet5p-6b

sourceHugging Facebsd-3-clauseupdated 2y agoView on Hugging Face
15likes298downloads
config.json188 linesDownload Raw Back to root
1{2  "_name_or_path": "codet5p-6b",3  "architectures": [4    "CodeT5pEncoderDecoderModel"5  ],6  "auto_map": {7    "AutoConfig": "configuration_codet5p.CodeT5pConfig",8    "AutoModel": "modeling_codet5p.CodeGenModel",9    "AutoModelForSeq2SeqLM": "modeling_codet5p.CodeT5pEncoderDecoderModel"10  },11  "decoder": {12    "_name_or_path": "codet5p-6b-decoder",13    "activation_function": "gelu_new",14    "add_cross_attention": true,15    "architectures": [16      "CodeT5pForCausalLM"17    ],18    "attn_pdrop": 0.0,19    "bad_words_ids": null,20    "bos_token_id": 1,21    "chunk_size_feed_forward": 0,22    "cross_attention_hidden_size": null,23    "decoder_start_token_id": null,24    "diversity_penalty": 0.0,25    "do_sample": false,26    "early_stopping": false,27    "embd_pdrop": 0.0,28    "encoder_no_repeat_ngram_size": 0,29    "eos_token_id": 50256,30    "exponential_decay_length_penalty": null,31    "finetuning_task": null,32    "forced_bos_token_id": null,33    "forced_eos_token_id": null,34    "gradient_checkpointing": false,35    "id2label": {36      "0": "LABEL_0",37      "1": "LABEL_1"38    },39    "initializer_range": 0.02,40    "is_decoder": true,41    "is_encoder_decoder": false,42    "label2id": {43      "LABEL_0": 0,44      "LABEL_1": 145    },46    "layer_norm_epsilon": 1e-05,47    "length_penalty": 1.0,48    "max_length": 20,49    "min_length": 0,50    "model_type": "codet5p_module",51    "n_ctx": 2048,52    "n_embd": 4096,53    "n_head": 16,54    "n_inner": null,55    "n_layer": 33,56    "n_positions": 2048,57    "no_repeat_ngram_size": 0,58    "num_beam_groups": 1,59    "num_beams": 1,60    "num_return_sequences": 1,61    "output_attentions": false,62    "output_hidden_states": false,63    "output_scores": false,64    "pad_token_id": null,65    "prefix": null,66    "problem_type": null,67    "pruned_heads": {},68    "remove_invalid_values": false,69    "repetition_penalty": 1.0,70    "resid_pdrop": 0.0,71    "return_dict": true,72    "return_dict_in_generate": false,73    "rotary_dim": 64,74    "scale_attn_weights": true,75    "sep_token_id": null,76    "summary_activation": null,77    "summary_first_dropout": 0.1,78    "summary_proj_to_labels": true,79    "summary_type": "cls_index",80    "summary_use_proj": true,81    "task_specific_params": null,82    "temperature": 1.0,83    "tf_legacy_loss": false,84    "tie_encoder_decoder": false,85    "tie_word_embeddings": false,86    "tokenizer_class": "GPT2Tokenizer",87    "top_k": 50,88    "top_p": 1.0,89    "torch_dtype": "float16",90    "torchscript": false,91    "transformers_version": "4.21.3",92    "typical_p": 1.0,93    "use_bfloat16": false,94    "use_cache": true,95    "vocab_size": 5120096  },97  "encoder": {98    "_name_or_path": "codet5p-350m-encoder",99    "activation_function": "gelu_new",100    "add_cross_attention": false,101    "architectures": [102      "CodeT5pModel"103    ],104    "attn_pdrop": 0.0,105    "bad_words_ids": null,106    "bos_token_id": 1,107    "chunk_size_feed_forward": 0,108    "cross_attention_hidden_size": null,109    "decoder_start_token_id": null,110    "diversity_penalty": 0.0,111    "do_sample": false,112    "early_stopping": false,113    "embd_pdrop": 0.0,114    "encoder_no_repeat_ngram_size": 0,115    "eos_token_id": 50256,116    "exponential_decay_length_penalty": null,117    "finetuning_task": null,118    "forced_bos_token_id": null,119    "forced_eos_token_id": null,120    "gradient_checkpointing": false,121    "id2label": {122      "0": "LABEL_0",123      "1": "LABEL_1"124    },125    "initializer_range": 0.02,126    "is_decoder": false,127    "is_encoder_decoder": false,128    "label2id": {129      "LABEL_0": 0,130      "LABEL_1": 1131    },132    "layer_norm_epsilon": 1e-05,133    "length_penalty": 1.0,134    "max_length": 20,135    "min_length": 0,136    "model_type": "codet5p_module",137    "n_ctx": 2048,138    "n_embd": 1024,139    "n_head": 16,140    "n_inner": null,141    "n_layer": 20,142    "n_positions": 2048,143    "no_repeat_ngram_size": 0,144    "num_beam_groups": 1,145    "num_beams": 1,146    "num_return_sequences": 1,147    "output_attentions": false,148    "output_hidden_states": false,149    "output_scores": false,150    "pad_token_id": null,151    "prefix": null,152    "problem_type": null,153    "pruned_heads": {},154    "remove_invalid_values": false,155    "repetition_penalty": 1.0,156    "resid_pdrop": 0.0,157    "return_dict": true,158    "return_dict_in_generate": false,159    "rotary_dim": 32,160    "scale_attn_weights": true,161    "sep_token_id": null,162    "summary_activation": null,163    "summary_first_dropout": 0.1,164    "summary_proj_to_labels": true,165    "summary_type": "cls_index",166    "summary_use_proj": true,167    "task_specific_params": null,168    "temperature": 1.0,169    "tf_legacy_loss": false,170    "tie_encoder_decoder": false,171    "tie_word_embeddings": false,172    "tokenizer_class": "GPT2Tokenizer",173    "top_k": 50,174    "top_p": 1.0,175    "torch_dtype": "float16",176    "torchscript": false,177    "transformers_version": "4.21.3",178    "typical_p": 1.0,179    "use_bfloat16": false,180    "use_cache": true,181    "vocab_size": 51200182  },183  "is_encoder_decoder": true,184  "model_type": "codet5p",185  "torch_dtype": "float16",186  "transformers_version": null187}188