hf-tiny-model-private/tiny-random-GPTNeoXModel
022
1{2 "architectures": [3 "GPTNeoXModel"4 ],5 "attention_probs_dropout_prob": 0.1,6 "bos_token_id": 0,7 "eos_token_id": 0,8 "hidden_act": "gelu",9 "hidden_dropout_prob": 0.1,10 "hidden_size": 32,11 "initializer_range": 0.02,12 "intermediate_size": 37,13 "layer_norm_eps": 1e-05,14 "max_position_embeddings": 512,15 "model_type": "gpt_neox",16 "num_attention_heads": 4,17 "num_hidden_layers": 5,18 "rotary_emb_base": 10000,19 "rotary_pct": 0.25,20 "tie_word_embeddings": false,21 "torch_dtype": "float32",22 "transformers_version": "4.28.0.dev0",23 "type_vocab_size": 16,24 "use_cache": true,25 "use_parallel_residual": true,26 "vocab_size": 102427}28 