bumblebee-testing/tiny-random-Phi3ForSequenceClassification
032
1{2 "architectures": [3 "Phi3ForSequenceClassification"4 ],5 "attention_dropout": 0.0,6 "attention_probs_dropout_prob": 0.1,7 "bos_token_id": 1,8 "embd_pdrop": 0.0,9 "eos_token_id": 32000,10 "hidden_act": "gelu",11 "hidden_dropout_prob": 0.1,12 "hidden_size": 32,13 "initializer_range": 0.02,14 "intermediate_size": 37,15 "max_position_embeddings": 512,16 "model_type": "phi3",17 "num_attention_heads": 4,18 "num_hidden_layers": 2,19 "num_key_value_heads": 4,20 "original_max_position_embeddings": 4096,21 "pad_token_id": 0,22 "resid_pdrop": 0.0,23 "rms_norm_eps": 1e-05,24 "rope_scaling": null,25 "rope_theta": 10000.0,26 "sliding_window": null,27 "tie_word_embeddings": false,28 "torch_dtype": "float32",29 "transformers_version": "4.41.0.dev0",30 "type_vocab_size": 16,31 "use_cache": true,32 "vocab_size": 102433}34 