{ "_name_or_path": "models/kl3m-doc-pico-long-001", "architectures": [ "RobertaModel" ], "attention_probs_dropout_prob": 0.1, "bos_token_id": 0, "classifier_dropout": null, "eos_token_id": 1, "hidden_act": "gelu_new", "hidden_dropout_prob": 0.1, "hidden_size": 256, "initializer_range": 0.02, "intermediate_size": 1024, "layer_norm_eps": 1e-08, "max_position_embeddings": 4096, "model_type": "roberta", "num_attention_heads": 8, "num_hidden_layers": 8, "pad_token_id": 2, "position_embedding_type": "absolute", "torch_dtype": "float32", "transformers_version": "4.49.0", "type_vocab_size": 1, "use_cache": true, "vocab_size": 131072 }