| { |
| "_name_or_path": "/mnt/quant_models/gemma2-2b-custom", |
| "architectures": [ |
| "Gemma2ForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "attn_logit_softcapping": 50.0, |
| "bos_token_id": 128000, |
| "cache_implementation": "hybrid", |
| "eos_token_id": 128001, |
| "final_logit_softcapping": 30.0, |
| "head_dim": 256, |
| "hidden_act": "gelu_pytorch_tanh", |
| "hidden_activation": "gelu_pytorch_tanh", |
| "hidden_size": 2304, |
| "initializer_range": 0.02, |
| "intermediate_size": 9216, |
| "max_position_embeddings": 8192, |
| "model_type": "gemma2", |
| "num_attention_heads": 8, |
| "num_hidden_layers": 26, |
| "num_key_value_heads": 4, |
| "pad_token_id": 0, |
| "quantization_config": { |
| "bits": 4, |
| "export": { |
| "kv_cache_group": [], |
| "pack_method": "reorder", |
| "weight_format": "real_quantized", |
| "weight_merge_groups": null |
| }, |
| "group_size": 128, |
| "modules_to_not_convert": [], |
| "pack_method": "reorder", |
| "quant_method": "awq", |
| "version": "gemm", |
| "zero_point": true |
| }, |
| "query_pre_attn_scalar": 256, |
| "rms_norm_eps": 1e-06, |
| "rope_theta": 10000.0, |
| "sliding_window": 4096, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.49.0", |
| "use_cache": false, |
| "vocab_size": 128256 |
| } |
|
|
