| { |
| "architectures": [ |
| "LlamaForCausalLM" |
| ], |
| "bos_token_id": 128000, |
| "eos_token_id": 128001, |
| "hidden_act": "silu", |
| "hidden_size": 2560, |
| "initializer_range": 0.02, |
| "intermediate_size": 6912, |
| "max_position_embeddings": 131072, |
| "model_type": "llama", |
| "rms_norm_eps": 1e-05, |
| "num_attention_heads": 20, |
| "num_hidden_layers": 30, |
| "num_key_value_heads": 5, |
| "rope_theta": 500000.0, |
| "tie_word_embeddings": true, |
| "torch_dtype": "bfloat16", |
| "use_cache": true, |
| "vocab_size": 128256, |
| "quantization_config": { |
| "config_groups": { |
| "group_0": { |
| "input_activations": { |
| "block_structure": null, |
| "dynamic": true, |
| "group_size": null, |
| "num_bits": 8, |
| "observer": "memoryless", |
| "observer_kwargs": {}, |
| "strategy": "token", |
| "symmetric": true, |
| "type": "float" |
| }, |
| "output_activations": null, |
| "targets": [ |
| "Linear" |
| ], |
| "weights": { |
| "block_structure": null, |
| "dynamic": false, |
| "group_size": null, |
| "num_bits": 8, |
| "observer": "minmax", |
| "observer_kwargs": {}, |
| "strategy": "tensor", |
| "symmetric": true, |
| "type": "float" |
| } |
| } |
| }, |
| "format": "float-quantized", |
| "ignore": [ |
| "lm_head" |
| ], |
| "kv_cache_scheme": null, |
| "quant_method": "compressed-tensors", |
| "quantization_status": "frozen" |
| }, |
| "head_dim": 128, |
| "transformers_version": "4.44.1" |
| } |
