| { |
| "architectures": [ |
| "LagunaForCausalLM" |
| ], |
| "auto_map": { |
| "AutoConfig": "configuration_laguna.LagunaConfig", |
| "AutoModelForCausalLM": "modeling_laguna.LagunaForCausalLM" |
| }, |
| "model_type": "laguna", |
| "vocab_size": 100352, |
| "hidden_size": 2048, |
| "intermediate_size": 8192, |
| "num_hidden_layers": 40, |
| "num_attention_heads": 48, |
| "num_key_value_heads": 8, |
| "head_dim": 128, |
| "max_position_embeddings": 131072, |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "rms_norm_eps": 1e-06, |
| "num_experts": 256, |
| "num_experts_per_tok": 8, |
| "moe_intermediate_size": 512, |
| "shared_expert_intermediate_size": 512, |
| "router_aux_loss_coef": 0.0, |
| "bos_token_id": 2, |
| "eos_token_id": [ |
| 2, |
| 24 |
| ], |
| "pad_token_id": 9, |
| "tie_word_embeddings": false, |
| "use_cache": true, |
| "torch_dtype": "bfloat16", |
| "gating": true, |
| "sliding_window": 512, |
| "rope_parameters": { |
| "full_attention": { |
| "rope_theta": 500000.0, |
| "rope_type": "yarn", |
| "factor": 32.0, |
| "original_max_position_embeddings": 4096, |
| "beta_slow": 1.0, |
| "beta_fast": 64.0, |
| "attention_factor": 1.0, |
| "partial_rotary_factor": 0.5 |
| }, |
| "sliding_attention": { |
| "rope_type": "default", |
| "rope_theta": 10000.0, |
| "partial_rotary_factor": 1.0 |
| }, |
| "original_max_position_embeddings": 4096 |
| }, |
| "layer_types": [ |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "full_attention", |
| "sliding_attention", |
| "sliding_attention", |
| "sliding_attention" |
| ], |
| "moe_apply_router_weight_on_input": false, |
| "partial_rotary_factor": 0.5, |
| "mlp_layer_types": [ |
| "dense", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse", |
| "sparse" |
| ], |
| "use_bidirectional_attention": false, |
| "moe_routed_scaling_factor": 2.5, |
| "num_attention_heads_per_layer": [ |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64, |
| 48, |
| 64, |
| 64, |
| 64 |
| ], |
| "quantization_config": { |
| "config_groups": { |
| "group_0": { |
| "format": "float-quantized", |
| "input_activations": { |
| "actorder": null, |
| "block_structure": null, |
| "dynamic": true, |
| "group_size": 128, |
| "num_bits": 8, |
| "observer": null, |
| "observer_kwargs": {}, |
| "scale_dtype": null, |
| "strategy": "group", |
| "symmetric": true, |
| "type": "float", |
| "zp_dtype": null |
| }, |
| "output_activations": null, |
| "targets": [ |
| "Linear" |
| ], |
| "weights": { |
| "actorder": null, |
| "block_structure": [ |
| 128, |
| 128 |
| ], |
| "dynamic": false, |
| "group_size": null, |
| "num_bits": 8, |
| "observer": "memoryless_minmax", |
| "observer_kwargs": {}, |
| "scale_dtype": null, |
| "strategy": "block", |
| "symmetric": true, |
| "type": "float", |
| "zp_dtype": null |
| } |
| } |
| }, |
| "format": "float-quantized", |
| "global_compression_ratio": null, |
| "ignore": [ |
| "lm_head", |
| "re:.*\\.self_attn\\.q_proj$", |
| "re:.*\\.self_attn\\.k_proj$", |
| "re:.*\\.self_attn\\.v_proj$", |
| "re:.*\\.self_attn\\.o_proj$", |
| "re:.*\\.self_attn\\.g_proj$", |
| "re:.*\\.mlp\\.gate$" |
| ], |
| "kv_cache_scheme": { |
| "actorder": null, |
| "block_structure": null, |
| "dynamic": false, |
| "group_size": null, |
| "num_bits": 8, |
| "observer": "minmax", |
| "observer_kwargs": {}, |
| "scale_dtype": null, |
| "strategy": "tensor", |
| "symmetric": true, |
| "type": "float", |
| "zp_dtype": null |
| }, |
| "quant_method": "compressed-tensors", |
| "quantization_status": "compressed", |
| "sparsity_config": {}, |
| "transform_config": { |
| "config_groups": { |
| "R1": { |
| "apply": [ |
| { |
| "ignore": [], |
| "inverse": false, |
| "location": "weight_output", |
| "targets": [ |
| "re:.*embed_tokens$", |
| "re:.*o_proj$", |
| "re:.*down_proj$" |
| ] |
| }, |
| { |
| "ignore": [], |
| "inverse": true, |
| "location": "weight_input", |
| "targets": [ |
| "re:.*q_proj$", |
| "re:.*k_proj$", |
| "re:.*v_proj$", |
| "re:.*gate_proj$", |
| "re:.*up_proj$", |
| "re:.*mlp.gate$", |
| "re:.*g_proj$", |
| "re:.*lm_head$" |
| ] |
| } |
| ], |
| "head_dim": 128, |
| "precision": "torch.float64", |
| "randomize": false, |
| "requires_grad": false, |
| "type": "hadamard" |
| } |
| } |
| }, |
| "version": "0.14.1.dev11+gf2ee47b" |
| } |
| } |
