| { |
| "architectures": [ |
| "Grok1ForCausalLM" |
| ], |
| "attn_factor": 1.0, |
| "attn_logit_softcapping": 30.0, |
| "attn_temperature_len": 1024, |
| "auto_map": { |
| "AutoConfig": "modeling_grok2.Grok2Config", |
| "AutoModelForCausalLM": "modeling_grok2.Grok1ForCausalLM" |
| }, |
| "beta_fast": 8, |
| "beta_slow": 1, |
| "embedding_multiplier_scale": 90.50966799187809, |
| "extrapolation_factor": 1.0, |
| "final_logit_softcapping": 50, |
| "global_attn_every_n": 1, |
| "head_dim": 128, |
| "hidden_size": 8192, |
| "intermediate_size": 32768, |
| "max_position_embeddings": 131072, |
| "model_type": "grok2", |
| "moe_intermediate_size": 16384, |
| "num_attention_heads": 64, |
| "num_experts_per_tok": 2, |
| "num_hidden_layers": 64, |
| "num_key_value_heads": 8, |
| "num_local_experts": 8, |
| "original_max_position_embeddings": 8192, |
| "output_multiplier_scale": 0.5, |
| "quantization_config": { |
| "config_groups": { |
| "FP8_DYNAMIC": { |
| "format": "float-quantized", |
| "input_activations": { |
| "actorder": null, |
| "block_structure": null, |
| "dynamic": true, |
| "group_size": null, |
| "num_bits": 8, |
| "observer": null, |
| "observer_kwargs": {}, |
| "scale_dtype": null, |
| "strategy": "token", |
| "symmetric": true, |
| "type": "float", |
| "zp_dtype": null |
| }, |
| "output_activations": null, |
| "targets": [ |
| "Linear" |
| ], |
| "weights": { |
| "actorder": null, |
| "block_structure": null, |
| "dynamic": false, |
| "group_size": null, |
| "num_bits": 8, |
| "observer": "memoryless_minmax", |
| "observer_kwargs": {}, |
| "scale_dtype": null, |
| "strategy": "channel", |
| "symmetric": true, |
| "type": "float", |
| "zp_dtype": null |
| } |
| } |
| }, |
| "format": "float-quantized", |
| "global_compression_ratio": null, |
| "ignore": [ |
| "lm_head", |
| "model.layers.40.block_sparse_moe.gate", |
| "model.layers.62.block_sparse_moe.gate", |
| "model.layers.30.block_sparse_moe.gate", |
| "model.layers.52.block_sparse_moe.gate", |
| "model.layers.15.block_sparse_moe.gate", |
| "model.layers.35.block_sparse_moe.gate", |
| "model.layers.50.block_sparse_moe.gate", |
| "model.layers.53.block_sparse_moe.gate", |
| "model.layers.29.block_sparse_moe.gate", |
| "model.layers.57.block_sparse_moe.gate", |
| "model.layers.42.block_sparse_moe.gate", |
| "model.layers.60.block_sparse_moe.gate", |
| "model.layers.31.block_sparse_moe.gate", |
| "model.layers.36.block_sparse_moe.gate", |
| "model.layers.48.block_sparse_moe.gate", |
| "model.layers.11.block_sparse_moe.gate", |
| "model.layers.19.block_sparse_moe.gate", |
| "model.layers.45.block_sparse_moe.gate", |
| "model.layers.7.block_sparse_moe.gate", |
| "model.layers.51.block_sparse_moe.gate", |
| "model.layers.56.block_sparse_moe.gate", |
| "model.layers.17.block_sparse_moe.gate", |
| "model.layers.49.block_sparse_moe.gate", |
| "model.layers.16.block_sparse_moe.gate", |
| "model.layers.6.block_sparse_moe.gate", |
| "model.layers.27.block_sparse_moe.gate", |
| "model.layers.61.block_sparse_moe.gate", |
| "model.layers.23.block_sparse_moe.gate", |
| "model.layers.63.block_sparse_moe.gate", |
| "model.layers.34.block_sparse_moe.gate", |
| "model.layers.22.block_sparse_moe.gate", |
| "model.layers.37.block_sparse_moe.gate", |
| "model.layers.39.block_sparse_moe.gate", |
| "model.layers.20.block_sparse_moe.gate", |
| "model.layers.32.block_sparse_moe.gate", |
| "model.layers.13.block_sparse_moe.gate", |
| "model.layers.58.block_sparse_moe.gate", |
| "model.layers.43.block_sparse_moe.gate", |
| "model.layers.14.block_sparse_moe.gate", |
| "model.layers.8.block_sparse_moe.gate", |
| "model.layers.18.block_sparse_moe.gate", |
| "model.layers.24.block_sparse_moe.gate", |
| "model.layers.54.block_sparse_moe.gate", |
| "model.layers.26.block_sparse_moe.gate", |
| "model.layers.33.block_sparse_moe.gate", |
| "model.layers.41.block_sparse_moe.gate", |
| "model.layers.38.block_sparse_moe.gate", |
| "model.layers.47.block_sparse_moe.gate", |
| "model.layers.9.block_sparse_moe.gate", |
| "model.layers.12.block_sparse_moe.gate", |
| "model.layers.10.block_sparse_moe.gate", |
| "model.layers.46.block_sparse_moe.gate", |
| "model.layers.25.block_sparse_moe.gate", |
| "model.layers.21.block_sparse_moe.gate", |
| "model.layers.28.block_sparse_moe.gate", |
| "model.layers.44.block_sparse_moe.gate", |
| "model.layers.59.block_sparse_moe.gate", |
| "model.embed_tokens", |
| "model.layers.55.block_sparse_moe.gate" |
| ], |
| "kv_cache_scheme": null, |
| "quant_method": "compressed-tensors", |
| "quantization_status": "compressed", |
| "sparsity_config": {}, |
| "transform_config": {}, |
| "version": "0.14.0.1" |
| }, |
| "residual_moe": true, |
| "rms_norm_eps": 1e-05, |
| "rope_theta": 208533496, |
| "rope_type": "original", |
| "router_logit_softcapping": 30.0, |
| "scaling_factor": 16.0, |
| "sliding_window_size": -1, |
| "torch_dtype": "bfloat16", |
| "vocab_size": 131072 |
| } |
