[2026-08-13 15:48:49,616] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:1747] baseline 0.000GB () [2026-08-13 15:48:49,617] [INFO] [axolotl.cli.config.load_cfg:333] [PID:1747] config: { "activation_offloading": false, "adapter": "lora", "attn_implementation": "flash_attention_2", "attn_needs_dtype_cast": true, "attn_supports_packing": true, "attn_uses_flash_lib": true, "axolotl_config_path": "/workspace/python4-aft-v2-20260813T154138Z-control/experiments/python4/aft_v2/runs/20260813T154138Z/arms/control/train/axolotl.yaml", "base_model": "/workspace/python4-aft-v2-state/20260813T154138Z/control/parent/control/sft/end", "base_model_config": "/workspace/python4-aft-v2-state/20260813T154138Z/control/parent/control/sft/end", "batch_size": 32, "bf16": true, "capabilities": { "bf16": true, "compute_capability": "sm_90", "fp8": true, "n_gpu": 1, "n_node": 1, "tf32": true }, "chat_template": "gemma3", "checkpoint_schedule": [ 128 ], "context_parallel_size": 1, "cosine_min_lr_ratio": 0.1, "dataloader_num_workers": 1, "dataloader_pin_memory": true, "dataloader_prefetch_factor": 256, "dataset_num_proc": 16, "dataset_prepared_path": "/workspace/python4-aft-v2-20260813T154138Z-control/experiments/python4/aft_v2/runs/20260813T154138Z/arms/control/train/prepared", "datasets": [ { "chat_template": "tokenizer_default", "field_messages": "messages", "message_property_mappings": { "content": "content", "role": "role" }, "path": "/workspace/python4-aft-v2-20260813T154138Z-control/experiments/python4/aft_v2/runs/20260813T154138Z/arms/control/aft_training.jsonl", "trust_remote_code": false, "type": "chat_template" } ], "ddp": false, "ddp_find_unused_parameters": true, "device": "cuda:0", "dion_rank_fraction": 1.0, "dion_rank_multiple_of": 1, "eaft_alpha": 1.0, "eaft_k": 20, "env_capabilities": { "torch_version": "2.12.1" }, "eot_tokens": [ "" ], "eval_batch_size": 4, "eval_causal_lm_metrics": [ "sacrebleu", "comet", "ter", "chrf" ], "eval_max_new_tokens": 128, "eval_table_size": 0, "experimental_skip_move_to_device": true, "fp16": false, "generate_samples": false, "generation_do_sample": true, "generation_max_new_tokens": 50, "generation_prompt_ratio": 0.5, "generation_temperature": 0.7, "gradient_accumulation_steps": 8, "gradient_checkpointing": true, "gradient_checkpointing_kwargs": { "use_reentrant": false }, "include_tkps": true, "is_multimodal": true, "layer_offloading": false, "learning_rate": 0.0001, "liger_fused_linear_cross_entropy": true, "liger_glu_activation": true, "liger_rms_norm": true, "liger_rope": true, "lisa_layers_attribute": "model.layers", "load_best_model_at_end": false, "load_in_4bit": false, "load_in_8bit": false, "local_rank": 0, "logging_steps": 1, "lora_alpha": 128, "lora_dropout": 0.0, "lora_embedding_kernel": true, "lora_mlp_kernel": true, "lora_o_kernel": true, "lora_qkv_kernel": true, "lora_r": 64, "lora_target_modules": [ "model.language_model.layers.0.self_attn.q_proj", "model.language_model.layers.0.self_attn.k_proj", "model.language_model.layers.0.self_attn.v_proj", "model.language_model.layers.0.self_attn.o_proj", "model.language_model.layers.0.mlp.gate_proj", "model.language_model.layers.0.mlp.up_proj", "model.language_model.layers.0.mlp.down_proj", "model.language_model.layers.1.self_attn.q_proj", "model.language_model.layers.1.self_attn.k_proj", "model.language_model.layers.1.self_attn.v_proj", "model.language_model.layers.1.self_attn.o_proj", "model.language_model.layers.1.mlp.gate_proj", "model.language_model.layers.1.mlp.up_proj", "model.language_model.layers.1.mlp.down_proj", "model.language_model.layers.2.self_attn.q_proj", "model.language_model.layers.2.self_attn.k_proj", "model.language_model.layers.2.self_attn.v_proj", "model.language_model.layers.2.self_attn.o_proj", "model.language_model.layers.2.mlp.gate_proj", "model.language_model.layers.2.mlp.up_proj", "model.language_model.layers.2.mlp.down_proj", "model.language_model.layers.3.self_attn.q_proj", "model.language_model.layers.3.self_attn.k_proj", "model.language_model.layers.3.self_attn.v_proj", "model.language_model.layers.3.self_attn.o_proj", "model.language_model.layers.3.mlp.gate_proj", "model.language_model.layers.3.mlp.up_proj", "model.language_model.layers.3.mlp.down_proj", "model.language_model.layers.4.self_attn.q_proj", "model.language_model.layers.4.self_attn.k_proj", "model.language_model.layers.4.self_attn.v_proj", "model.language_model.layers.4.self_attn.o_proj", "model.language_model.layers.4.mlp.gate_proj", "model.language_model.layers.4.mlp.up_proj", "model.language_model.layers.4.mlp.down_proj", "model.language_model.layers.5.self_attn.q_proj", "model.language_model.layers.5.self_attn.k_proj", "model.language_model.layers.5.self_attn.v_proj", "model.language_model.layers.5.self_attn.o_proj", "model.language_model.layers.5.mlp.gate_proj", "model.language_model.layers.5.mlp.up_proj", "model.language_model.layers.5.mlp.down_proj", "model.language_model.layers.6.self_attn.q_proj", "model.language_model.layers.6.self_attn.k_proj", "model.language_model.layers.6.self_attn.v_proj", "model.language_model.layers.6.self_attn.o_proj", "model.language_model.layers.6.mlp.gate_proj", "model.language_model.layers.6.mlp.up_proj", "model.language_model.layers.6.mlp.down_proj", "model.language_model.layers.7.self_attn.q_proj", "model.language_model.layers.7.self_attn.k_proj", "model.language_model.layers.7.self_attn.v_proj", "model.language_model.layers.7.self_attn.o_proj", "model.language_model.layers.7.mlp.gate_proj", "model.language_model.layers.7.mlp.up_proj", "model.language_model.layers.7.mlp.down_proj", "model.language_model.layers.8.self_attn.q_proj", "model.language_model.layers.8.self_attn.k_proj", "model.language_model.layers.8.self_attn.v_proj", "model.language_model.layers.8.self_attn.o_proj", "model.language_model.layers.8.mlp.gate_proj", "model.language_model.layers.8.mlp.up_proj", "model.language_model.layers.8.mlp.down_proj", "model.language_model.layers.9.self_attn.q_proj", "model.language_model.layers.9.self_attn.k_proj", "model.language_model.layers.9.self_attn.v_proj", "model.language_model.layers.9.self_attn.o_proj", "model.language_model.layers.9.mlp.gate_proj", "model.language_model.layers.9.mlp.up_proj", "model.language_model.layers.9.mlp.down_proj", "model.language_model.layers.10.self_attn.q_proj", "model.language_model.layers.10.self_attn.k_proj", "model.language_model.layers.10.self_attn.v_proj", "model.language_model.layers.10.self_attn.o_proj", "model.language_model.layers.10.mlp.gate_proj", "model.language_model.layers.10.mlp.up_proj", "model.language_model.layers.10.mlp.down_proj", "model.language_model.layers.11.self_attn.q_proj", "model.language_model.layers.11.self_attn.k_proj", "model.language_model.layers.11.self_attn.v_proj", "model.language_model.layers.11.self_attn.o_proj", "model.language_model.layers.11.mlp.gate_proj", "model.language_model.layers.11.mlp.up_proj", "model.language_model.layers.11.mlp.down_proj", "model.language_model.layers.12.self_attn.q_proj", "model.language_model.layers.12.self_attn.k_proj", "model.language_model.layers.12.self_attn.v_proj", "model.language_model.layers.12.self_attn.o_proj", "model.language_model.layers.12.mlp.gate_proj", "model.language_model.layers.12.mlp.up_proj", "model.language_model.layers.12.mlp.down_proj", "model.language_model.layers.13.self_attn.q_proj", "model.language_model.layers.13.self_attn.k_proj", "model.language_model.layers.13.self_attn.v_proj", "model.language_model.layers.13.self_attn.o_proj", "model.language_model.layers.13.mlp.gate_proj", "model.language_model.layers.13.mlp.up_proj", "model.language_model.layers.13.mlp.down_proj", "model.language_model.layers.14.self_attn.q_proj", "model.language_model.layers.14.self_attn.k_proj", "model.language_model.layers.14.self_attn.v_proj", "model.language_model.layers.14.self_attn.o_proj", "model.language_model.layers.14.mlp.gate_proj", "model.language_model.layers.14.mlp.up_proj", "model.language_model.layers.14.mlp.down_proj", "model.language_model.layers.15.self_attn.q_proj", "model.language_model.layers.15.self_attn.k_proj", "model.language_model.layers.15.self_attn.v_proj", "model.language_model.layers.15.self_attn.o_proj", "model.language_model.layers.15.mlp.gate_proj", "model.language_model.layers.15.mlp.up_proj", "model.language_model.layers.15.mlp.down_proj", "model.language_model.layers.16.self_attn.q_proj", "model.language_model.layers.16.self_attn.k_proj", "model.language_model.layers.16.self_attn.v_proj", "model.language_model.layers.16.self_attn.o_proj", "model.language_model.layers.16.mlp.gate_proj", "model.language_model.layers.16.mlp.up_proj", "model.language_model.layers.16.mlp.down_proj", "model.language_model.layers.17.self_attn.q_proj", "model.language_model.layers.17.self_attn.k_proj", "model.language_model.layers.17.self_attn.v_proj", "model.language_model.layers.17.self_attn.o_proj", "model.language_model.layers.17.mlp.gate_proj", "model.language_model.layers.17.mlp.up_proj", "model.language_model.layers.17.mlp.down_proj", "model.language_model.layers.18.self_attn.q_proj", "model.language_model.layers.18.self_attn.k_proj", "model.language_model.layers.18.self_attn.v_proj", "model.language_model.layers.18.self_attn.o_proj", "model.language_model.layers.18.mlp.gate_proj", "model.language_model.layers.18.mlp.up_proj", "model.language_model.layers.18.mlp.down_proj", "model.language_model.layers.19.self_attn.q_proj", "model.language_model.layers.19.self_attn.k_proj", "model.language_model.layers.19.self_attn.v_proj", "model.language_model.layers.19.self_attn.o_proj", "model.language_model.layers.19.mlp.gate_proj", "model.language_model.layers.19.mlp.up_proj", "model.language_model.layers.19.mlp.down_proj", "model.language_model.layers.20.self_attn.q_proj", "model.language_model.layers.20.self_attn.k_proj", "model.language_model.layers.20.self_attn.v_proj", "model.language_model.layers.20.self_attn.o_proj", "model.language_model.layers.20.mlp.gate_proj", "model.language_model.layers.20.mlp.up_proj", "model.language_model.layers.20.mlp.down_proj", "model.language_model.layers.21.self_attn.q_proj", "model.language_model.layers.21.self_attn.k_proj", "model.language_model.layers.21.self_attn.v_proj", "model.language_model.layers.21.self_attn.o_proj", "model.language_model.layers.21.mlp.gate_proj", "model.language_model.layers.21.mlp.up_proj", "model.language_model.layers.21.mlp.down_proj", "model.language_model.layers.22.self_attn.q_proj", "model.language_model.layers.22.self_attn.k_proj", "model.language_model.layers.22.self_attn.v_proj", "model.language_model.layers.22.self_attn.o_proj", "model.language_model.layers.22.mlp.gate_proj", "model.language_model.layers.22.mlp.up_proj", "model.language_model.layers.22.mlp.down_proj", "model.language_model.layers.23.self_attn.q_proj", "model.language_model.layers.23.self_attn.k_proj", "model.language_model.layers.23.self_attn.v_proj", "model.language_model.layers.23.self_attn.o_proj", "model.language_model.layers.23.mlp.gate_proj", "model.language_model.layers.23.mlp.up_proj", "model.language_model.layers.23.mlp.down_proj", "model.language_model.layers.24.self_attn.q_proj", "model.language_model.layers.24.self_attn.k_proj", "model.language_model.layers.24.self_attn.v_proj", "model.language_model.layers.24.self_attn.o_proj", "model.language_model.layers.24.mlp.gate_proj", "model.language_model.layers.24.mlp.up_proj", "model.language_model.layers.24.mlp.down_proj", "model.language_model.layers.25.self_attn.q_proj", "model.language_model.layers.25.self_attn.k_proj", "model.language_model.layers.25.self_attn.v_proj", "model.language_model.layers.25.self_attn.o_proj", "model.language_model.layers.25.mlp.gate_proj", "model.language_model.layers.25.mlp.up_proj", "model.language_model.layers.25.mlp.down_proj", "model.language_model.layers.26.self_attn.q_proj", "model.language_model.layers.26.self_attn.k_proj", "model.language_model.layers.26.self_attn.v_proj", "model.language_model.layers.26.self_attn.o_proj", "model.language_model.layers.26.mlp.gate_proj", "model.language_model.layers.26.mlp.up_proj", "model.language_model.layers.26.mlp.down_proj", "model.language_model.layers.27.self_attn.q_proj", "model.language_model.layers.27.self_attn.k_proj", "model.language_model.layers.27.self_attn.v_proj", "model.language_model.layers.27.self_attn.o_proj", "model.language_model.layers.27.mlp.gate_proj", "model.language_model.layers.27.mlp.up_proj", "model.language_model.layers.27.mlp.down_proj", "model.language_model.layers.28.self_attn.q_proj", "model.language_model.layers.28.self_attn.k_proj", "model.language_model.layers.28.self_attn.v_proj", "model.language_model.layers.28.self_attn.o_proj", "model.language_model.layers.28.mlp.gate_proj", "model.language_model.layers.28.mlp.up_proj", "model.language_model.layers.28.mlp.down_proj", "model.language_model.layers.29.self_attn.q_proj", "model.language_model.layers.29.self_attn.k_proj", "model.language_model.layers.29.self_attn.v_proj", "model.language_model.layers.29.self_attn.o_proj", "model.language_model.layers.29.mlp.gate_proj", "model.language_model.layers.29.mlp.up_proj", "model.language_model.layers.29.mlp.down_proj", "model.language_model.layers.30.self_attn.q_proj", "model.language_model.layers.30.self_attn.k_proj", "model.language_model.layers.30.self_attn.v_proj", "model.language_model.layers.30.self_attn.o_proj", "model.language_model.layers.30.mlp.gate_proj", "model.language_model.layers.30.mlp.up_proj", "model.language_model.layers.30.mlp.down_proj", "model.language_model.layers.31.self_attn.q_proj", "model.language_model.layers.31.self_attn.k_proj", "model.language_model.layers.31.self_attn.v_proj", "model.language_model.layers.31.self_attn.o_proj", "model.language_model.layers.31.mlp.gate_proj", "model.language_model.layers.31.mlp.up_proj", "model.language_model.layers.31.mlp.down_proj", "model.language_model.layers.32.self_attn.q_proj", "model.language_model.layers.32.self_attn.k_proj", "model.language_model.layers.32.self_attn.v_proj", "model.language_model.layers.32.self_attn.o_proj", "model.language_model.layers.32.mlp.gate_proj", "model.language_model.layers.32.mlp.up_proj", "model.language_model.layers.32.mlp.down_proj", "model.language_model.layers.33.self_attn.q_proj", "model.language_model.layers.33.self_attn.k_proj", "model.language_model.layers.33.self_attn.v_proj", "model.language_model.layers.33.self_attn.o_proj", "model.language_model.layers.33.mlp.gate_proj", "model.language_model.layers.33.mlp.up_proj", "model.language_model.layers.33.mlp.down_proj", "model.language_model.layers.34.self_attn.q_proj", "model.language_model.layers.34.self_attn.k_proj", "model.language_model.layers.34.self_attn.v_proj", "model.language_model.layers.34.self_attn.o_proj", "model.language_model.layers.34.mlp.gate_proj", "model.language_model.layers.34.mlp.up_proj", "model.language_model.layers.34.mlp.down_proj", "model.language_model.layers.35.self_attn.q_proj", "model.language_model.layers.35.self_attn.k_proj", "model.language_model.layers.35.self_attn.v_proj", "model.language_model.layers.35.self_attn.o_proj", "model.language_model.layers.35.mlp.gate_proj", "model.language_model.layers.35.mlp.up_proj", "model.language_model.layers.35.mlp.down_proj", "model.language_model.layers.36.self_attn.q_proj", "model.language_model.layers.36.self_attn.k_proj", "model.language_model.layers.36.self_attn.v_proj", "model.language_model.layers.36.self_attn.o_proj", "model.language_model.layers.36.mlp.gate_proj", "model.language_model.layers.36.mlp.up_proj", "model.language_model.layers.36.mlp.down_proj", "model.language_model.layers.37.self_attn.q_proj", "model.language_model.layers.37.self_attn.k_proj", "model.language_model.layers.37.self_attn.v_proj", "model.language_model.layers.37.self_attn.o_proj", "model.language_model.layers.37.mlp.gate_proj", "model.language_model.layers.37.mlp.up_proj", "model.language_model.layers.37.mlp.down_proj", "model.language_model.layers.38.self_attn.q_proj", "model.language_model.layers.38.self_attn.k_proj", "model.language_model.layers.38.self_attn.v_proj", "model.language_model.layers.38.self_attn.o_proj", "model.language_model.layers.38.mlp.gate_proj", "model.language_model.layers.38.mlp.up_proj", "model.language_model.layers.38.mlp.down_proj", "model.language_model.layers.39.self_attn.q_proj", "model.language_model.layers.39.self_attn.k_proj", "model.language_model.layers.39.self_attn.v_proj", "model.language_model.layers.39.self_attn.o_proj", "model.language_model.layers.39.mlp.gate_proj", "model.language_model.layers.39.mlp.up_proj", "model.language_model.layers.39.mlp.down_proj", "model.language_model.layers.40.self_attn.q_proj", "model.language_model.layers.40.self_attn.k_proj", "model.language_model.layers.40.self_attn.v_proj", "model.language_model.layers.40.self_attn.o_proj", "model.language_model.layers.40.mlp.gate_proj", "model.language_model.layers.40.mlp.up_proj", "model.language_model.layers.40.mlp.down_proj", "model.language_model.layers.41.self_attn.q_proj", "model.language_model.layers.41.self_attn.k_proj", "model.language_model.layers.41.self_attn.v_proj", "model.language_model.layers.41.self_attn.o_proj", "model.language_model.layers.41.mlp.gate_proj", "model.language_model.layers.41.mlp.up_proj", "model.language_model.layers.41.mlp.down_proj", "model.language_model.layers.42.self_attn.q_proj", "model.language_model.layers.42.self_attn.k_proj", "model.language_model.layers.42.self_attn.v_proj", "model.language_model.layers.42.self_attn.o_proj", "model.language_model.layers.42.mlp.gate_proj", "model.language_model.layers.42.mlp.up_proj", "model.language_model.layers.42.mlp.down_proj", "model.language_model.layers.43.self_attn.q_proj", "model.language_model.layers.43.self_attn.k_proj", "model.language_model.layers.43.self_attn.v_proj", "model.language_model.layers.43.self_attn.o_proj", "model.language_model.layers.43.mlp.gate_proj", "model.language_model.layers.43.mlp.up_proj", "model.language_model.layers.43.mlp.down_proj", "model.language_model.layers.44.self_attn.q_proj", "model.language_model.layers.44.self_attn.k_proj", "model.language_model.layers.44.self_attn.v_proj", "model.language_model.layers.44.self_attn.o_proj", "model.language_model.layers.44.mlp.gate_proj", "model.language_model.layers.44.mlp.up_proj", "model.language_model.layers.44.mlp.down_proj", "model.language_model.layers.45.self_attn.q_proj", "model.language_model.layers.45.self_attn.k_proj", "model.language_model.layers.45.self_attn.v_proj", "model.language_model.layers.45.self_attn.o_proj", "model.language_model.layers.45.mlp.gate_proj", "model.language_model.layers.45.mlp.up_proj", "model.language_model.layers.45.mlp.down_proj", "model.language_model.layers.46.self_attn.q_proj", "model.language_model.layers.46.self_attn.k_proj", "model.language_model.layers.46.self_attn.v_proj", "model.language_model.layers.46.self_attn.o_proj", "model.language_model.layers.46.mlp.gate_proj", "model.language_model.layers.46.mlp.up_proj", "model.language_model.layers.46.mlp.down_proj", "model.language_model.layers.47.self_attn.q_proj", "model.language_model.layers.47.self_attn.k_proj", "model.language_model.layers.47.self_attn.v_proj", "model.language_model.layers.47.self_attn.o_proj", "model.language_model.layers.47.mlp.gate_proj", "model.language_model.layers.47.mlp.up_proj", "model.language_model.layers.47.mlp.down_proj", "model.language_model.layers.48.self_attn.q_proj", "model.language_model.layers.48.self_attn.k_proj", "model.language_model.layers.48.self_attn.v_proj", "model.language_model.layers.48.self_attn.o_proj", "model.language_model.layers.48.mlp.gate_proj", "model.language_model.layers.48.mlp.up_proj", "model.language_model.layers.48.mlp.down_proj", "model.language_model.layers.49.self_attn.q_proj", "model.language_model.layers.49.self_attn.k_proj", "model.language_model.layers.49.self_attn.v_proj", "model.language_model.layers.49.self_attn.o_proj", "model.language_model.layers.49.mlp.gate_proj", "model.language_model.layers.49.mlp.up_proj", "model.language_model.layers.49.mlp.down_proj", "model.language_model.layers.50.self_attn.q_proj", "model.language_model.layers.50.self_attn.k_proj", "model.language_model.layers.50.self_attn.v_proj", "model.language_model.layers.50.self_attn.o_proj", "model.language_model.layers.50.mlp.gate_proj", "model.language_model.layers.50.mlp.up_proj", "model.language_model.layers.50.mlp.down_proj", "model.language_model.layers.51.self_attn.q_proj", "model.language_model.layers.51.self_attn.k_proj", "model.language_model.layers.51.self_attn.v_proj", "model.language_model.layers.51.self_attn.o_proj", "model.language_model.layers.51.mlp.gate_proj", "model.language_model.layers.51.mlp.up_proj", "model.language_model.layers.51.mlp.down_proj", "model.language_model.layers.52.self_attn.q_proj", "model.language_model.layers.52.self_attn.k_proj", "model.language_model.layers.52.self_attn.v_proj", "model.language_model.layers.52.self_attn.o_proj", "model.language_model.layers.52.mlp.gate_proj", "model.language_model.layers.52.mlp.up_proj", "model.language_model.layers.52.mlp.down_proj", "model.language_model.layers.53.self_attn.q_proj", "model.language_model.layers.53.self_attn.k_proj", "model.language_model.layers.53.self_attn.v_proj", "model.language_model.layers.53.self_attn.o_proj", "model.language_model.layers.53.mlp.gate_proj", "model.language_model.layers.53.mlp.up_proj", "model.language_model.layers.53.mlp.down_proj", "model.language_model.layers.54.self_attn.q_proj", "model.language_model.layers.54.self_attn.k_proj", "model.language_model.layers.54.self_attn.v_proj", "model.language_model.layers.54.self_attn.o_proj", "model.language_model.layers.54.mlp.gate_proj", "model.language_model.layers.54.mlp.up_proj", "model.language_model.layers.54.mlp.down_proj", "model.language_model.layers.55.self_attn.q_proj", "model.language_model.layers.55.self_attn.k_proj", "model.language_model.layers.55.self_attn.v_proj", "model.language_model.layers.55.self_attn.o_proj", "model.language_model.layers.55.mlp.gate_proj", "model.language_model.layers.55.mlp.up_proj", "model.language_model.layers.55.mlp.down_proj", "model.language_model.layers.56.self_attn.q_proj", "model.language_model.layers.56.self_attn.k_proj", "model.language_model.layers.56.self_attn.v_proj", "model.language_model.layers.56.self_attn.o_proj", "model.language_model.layers.56.mlp.gate_proj", "model.language_model.layers.56.mlp.up_proj", "model.language_model.layers.56.mlp.down_proj", "model.language_model.layers.57.self_attn.q_proj", "model.language_model.layers.57.self_attn.k_proj", "model.language_model.layers.57.self_attn.v_proj", "model.language_model.layers.57.self_attn.o_proj", "model.language_model.layers.57.mlp.gate_proj", "model.language_model.layers.57.mlp.up_proj", "model.language_model.layers.57.mlp.down_proj", "model.language_model.layers.58.self_attn.q_proj", "model.language_model.layers.58.self_attn.k_proj", "model.language_model.layers.58.self_attn.v_proj", "model.language_model.layers.58.self_attn.o_proj", "model.language_model.layers.58.mlp.gate_proj", "model.language_model.layers.58.mlp.up_proj", "model.language_model.layers.58.mlp.down_proj", "model.language_model.layers.59.self_attn.q_proj", "model.language_model.layers.59.self_attn.k_proj", "model.language_model.layers.59.self_attn.v_proj", "model.language_model.layers.59.self_attn.o_proj", "model.language_model.layers.59.mlp.gate_proj", "model.language_model.layers.59.mlp.up_proj", "model.language_model.layers.59.mlp.down_proj", "model.language_model.layers.60.self_attn.q_proj", "model.language_model.layers.60.self_attn.k_proj", "model.language_model.layers.60.self_attn.v_proj", "model.language_model.layers.60.self_attn.o_proj", "model.language_model.layers.60.mlp.gate_proj", "model.language_model.layers.60.mlp.up_proj", "model.language_model.layers.60.mlp.down_proj", "model.language_model.layers.61.self_attn.q_proj", "model.language_model.layers.61.self_attn.k_proj", "model.language_model.layers.61.self_attn.v_proj", "model.language_model.layers.61.self_attn.o_proj", "model.language_model.layers.61.mlp.gate_proj", "model.language_model.layers.61.mlp.up_proj", "model.language_model.layers.61.mlp.down_proj" ], "loraplus_lr_embedding": 1e-06, "lr_scheduler": "cosine", "max_grad_norm": 1.0, "mean_resizing_embeddings": false, "merge_method": "memory_efficient", "micro_batch_size": 4, "model_config_type": "gemma3", "model_config_type_text": "gemma3_text", "num_epochs": 4.0, "num_generation_samples": 3, "optimizer": "adamw_torch_fused", "otel_metrics_host": "localhost", "otel_metrics_port": 8000, "output_dir": "/workspace/python4-aft-v2-20260813T154138Z-control/experiments/python4/aft_v2/runs/20260813T154138Z/arms/control/train/checkpoints", "pad_to_sequence_len": false, "plugins": [ "axolotl.integrations.liger.LigerPlugin", "scimt.train.axolotl_plugins.CheckpointSchedulePlugin" ], "pretrain_multipack_attn": true, "processor_config": "/workspace/python4-aft-v2-state/20260813T154138Z/control/parent/control/sft/end", "profiler_steps_start": 0, "qgalore_cos_threshold": 0.4, "qgalore_gamma_proj": 2, "qgalore_proj_bits": 4, "qgalore_proj_group_size": 256, "qgalore_proj_quant": true, "qgalore_proj_type": "std", "qgalore_queue_size": 5, "qgalore_rank": 256, "qgalore_scale": 0.25, "qgalore_update_proj_gap": 200, "qlora_sharded_model_loading": false, "quantize_moe_experts": false, "ray_num_workers": 1, "relora_prune_method": "magnitude", "resources_per_worker": { "GPU": 1 }, "sample_packing": false, "sample_packing_bin_size": 200, "sample_packing_group_size": 100000, "save_only_model": true, "save_safetensors": true, "save_strategy": "no", "save_total_limit": 1, "seed": 424242, "sequence_len": 4096, "shuffle_before_merging_datasets": false, "shuffle_merged_datasets": true, "skip_prepare_dataset": false, "streaming_multipack_buffer_size": 10000, "strict": false, "tensor_parallel_size": 1, "tf32": true, "tiled_mlp_use_original_mlp": true, "tokenizer_config": "/workspace/python4-aft-v2-state/20260813T154138Z/control/parent/control/sft/end", "tokenizer_save_jinja_files": true, "torch_dtype": "torch.bfloat16", "train_on_inputs": false, "trl": { "async_prefetch": false, "log_completions": false, "mask_truncated_completions": false, "ref_model_mixup_alpha": 0.9, "ref_model_sync_steps": 64, "replay_buffer_size": 0, "replay_recompute_logps": true, "reroll_max_groups": 1, "reroll_start_fraction": 1.0, "reward_num_workers": 1, "scale_rewards": true, "skip_zero_advantage_batches": true, "sync_ref_model": false, "use_data_producer": false, "use_vllm": false, "vllm_lora_sync": false, "vllm_server_host": "0.0.0.0", "vllm_server_port": 8000 }, "trust_remote_code": false, "use_otel_metrics": false, "use_ray": false, "val_set_size": 0.0, "vllm": { "device": "auto", "dtype": "auto", "gpu_memory_utilization": 0.9, "host": "0.0.0.0", "port": 8000 }, "warmup_ratio": 0.05, "weight_decay": 0.01, "world_size": 1 } [2026-08-13 15:48:50,617] [DEBUG] [axolotl.loaders.utils.check_model_config:88] [PID:1747] Loaded image size: 896 from model config [2026-08-13 15:48:52,258] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:311] [PID:1747] EOS: 106 / [2026-08-13 15:48:52,258] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:312] [PID:1747] BOS: 2 / [2026-08-13 15:48:52,258] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:313] [PID:1747] PAD: 0 / [2026-08-13 15:48:52,258] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:314] [PID:1747] UNK: 3 / [2026-08-13 15:48:52,259] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:482] [PID:1747] Unable to find prepared dataset in /workspace/python4-aft-v2-20260813T154138Z-control/experiments/python4/aft_v2/runs/20260813T154138Z/arms/control/train/prepared/cadb634638434a88a5119b219d5e7b61 [2026-08-13 15:48:52,259] [INFO] [axolotl.utils.data.sft._load_raw_datasets:320] [PID:1747] Loading raw datasets... [2026-08-13 15:48:52,259] [WARNING] [axolotl.utils.data.sft._load_raw_datasets:322] [PID:1747] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset using `axolotl preprocess path/to/config.yml`. Generating train split: 0 examples [00:00, ? examples/s] Generating train split: 1024 examples [00:00, 171750.60 examples/s] [2026-08-13 15:48:52,729] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:1747] Loading dataset: /workspace/python4-aft-v2-20260813T154138Z-control/experiments/python4/aft_v2/runs/20260813T154138Z/arms/control/aft_training.jsonl with base_type: chat_template and prompt_style: None [2026-08-13 15:48:52,736] [INFO] [axolotl.prompt_strategies.chat_template.__call__:1209] [PID:1747] Using chat template: --- {{ bos_token }} {%- if messages[0]['role'] == 'system' -%} {%- if messages[0]['content'] is string -%} {%- set first_user_prefix = messages[0]['content'] + ' ' -%} {%- else -%} {%- set first_user_prefix = messages[0]['content'][0]['text'] + ' ' -%} {%- endif -%} {%- set loop_messages = messages[1:] -%} {%- else -%} {%- set first_user_prefix = "" -%} {%- set loop_messages = messages -%} {%- endif -%} {%- for message in loop_messages -%} {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%} {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }} {%- endif -%} {%- if (message['role'] == 'assistant') -%} {%- set role = "model" -%} {%- else -%} {%- set role = message['role'] -%} {%- endif -%} {{ '' + role + ' ' + (first_user_prefix if loop.first else "") }} {%- if message['content'] is string -%} {{ message['content'] | trim }} {%- elif message['content'] is iterable -%} {%- for item in message['content'] -%} {%- if item['type'] == 'image' -%} {{ '' }} {%- elif item['type'] == 'text' -%} {{ item['text'] | trim }} {%- endif -%} {%- endfor -%} {%- else -%} {{ raise_exception("Invalid content type") }} {%- endif -%} {{ ' ' }} {%- endfor -%} {%- if add_generation_prompt -%} {{'model '}} {%- endif -%} --- Tokenizing Prompts (num_proc=16): 0%| | 0/1024 [00:004096) (num_proc=16): 0%| | 0/1024 [00:004096) (num_proc=16): 6%|▋ | 64/1024 [00:00<00:05, 189.02 examples/s] Dropping Invalid Sequences (4096) (num_proc=16): 100%|██████████| 1024/1024 [00:00<00:00, 1687.50 examples/s] Saving the dataset (0/4 shards): 0%| | 0/1024 [00:00 [2026-08-13 15:50:37,630] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:312] [PID:1747] BOS: 2 / [2026-08-13 15:50:37,630] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:313] [PID:1747] PAD: 0 / [2026-08-13 15:50:37,630] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:314] [PID:1747] UNK: 3 / [2026-08-13 15:50:39,540] [DEBUG] [axolotl.train.setup_model_and_tokenizer:81] [PID:1747] Loading model [2026-08-13 15:50:39,545] [DEBUG] [axolotl.monkeypatch.torchao_optim.patch_torchao_optim_state_8bit:75] [PID:1747] Patched OptimState8bit for torch.compile compatibility [2026-08-13 15:50:39,545] [DEBUG] [axolotl.monkeypatch.torchao_optim.patch_torchao_optim_state_8bit:122] [PID:1747] Patched OptimState4bit for torch.compile compatibility [2026-08-13 15:50:39,545] [DEBUG] [axolotl.monkeypatch.torchao_optim.patch_torchao_optim_state_8bit:154] [PID:1747] Patched OptimStateFp8 for torch.compile compatibility [2026-08-13 15:50:39,550] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:94] [PID:1747] Patched Trainer.evaluation_loop with nanmean loss calculation [2026-08-13 15:50:39,551] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:148] [PID:1747] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation [2026-08-13 15:50:39,621] [INFO] [axolotl.monkeypatch.attention.flash_attn_4.patch_flash_attn_4:52] [PID:1747] Flash Attention 4 is available for your GPU and offers faster training speeds. To enable: pip install flash-attn-4 [2026-08-13 15:50:40,840] [INFO] [axolotl.monkeypatch.lora_kernels.patch_self_attn_lora:304] [PID:1747] Patched attention class with LoRA optims: Gemma3Attention [2026-08-13 15:50:41,076] [INFO] [axolotl.integrations.liger.plugin.pre_model_load:117] [PID:1747] Applying LIGER to gemma3 with kwargs: {'rope': True, 'cross_entropy': None, 'fused_linear_cross_entropy': True, 'rms_norm': True, 'layer_norm': None, 'geglu': True} Loading weights: 0%| | 0/1248 [00:00