| { |
| "_name_or_path": "iGeniusAI/Italia-9B-Instruct-v0.1", |
| "architectures": [ |
| "GPTNeoXModel" |
| ], |
| "attention_bias": true, |
| "attention_dropout": 0.0, |
| "attention_probs_dropout_prob": 0, |
| "auto_map": { |
| "AutoConfig": "iGeniusAI/Italia-9B-Instruct-v0.1--configuration_italia.ItaliaConfig", |
| "AutoModel": "iGeniusAI/Italia-9B-Instruct-v0.1--modeling_italia.GPTNeoXModel", |
| "AutoModelForCausalLM": "iGeniusAI/Italia-9B-Instruct-v0.1--modeling_italia.ItaliaForCausalLM" |
| }, |
| "bos_token_id": 1, |
| "classifier_dropout": 0.1, |
| "eos_token_id": 2, |
| "hidden_act": "gelu_new", |
| "hidden_dropout": 0.0, |
| "hidden_size": 5120, |
| "initializer_range": 0.01, |
| "intermediate_size": 12800, |
| "layer_norm_eps": 1e-05, |
| "max_position_embeddings": 4096, |
| "model_type": "gpt_neox", |
| "num_attention_heads": 32, |
| "num_hidden_layers": 34, |
| "partial_rotary_factor": 0.4, |
| "quantization_config": { |
| "amp": false, |
| "autoround_version": "0.4.6", |
| "backend": "auto_round:gptq:exllamav2", |
| "batch_size": 4, |
| "bits": 4, |
| "data_type": "int", |
| "dataset": "NeelNanda/pile-10k", |
| "enable_minmax_tuning": true, |
| "enable_norm_bias_tuning": false, |
| "enable_quanted_input": true, |
| "gradient_accumulate_steps": 1, |
| "group_size": 64, |
| "iters": 200, |
| "low_gpu_mem_usage": false, |
| "lr": 0.005, |
| "minmax_lr": 0.005, |
| "nsamples": 128, |
| "quant_method": "intel/auto-round", |
| "scale_dtype": "torch.float16", |
| "seqlen": 512, |
| "sym": true, |
| "to_quant_block_names": null |
| }, |
| "rope_scaling": null, |
| "rope_theta": 10000, |
| "rotary_emb_base": 10000, |
| "rotary_pct": 0.4, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float16", |
| "transformers_version": "4.49.0", |
| "unk_token_id": 0, |
| "use_cache": false, |
| "use_parallel_residual": true, |
| "vocab_size": 50176 |
| } |
