[2026-05-20 01:49:25,315] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:100066] baseline 0.000GB () [2026-05-20 01:49:25,315] [INFO] [axolotl.cli.config.load_cfg:341] [PID:100066] config: { "activation_offloading": false, "adapter": "qlora", "attn_implementation": "flash_attention_2", "axolotl_config_path": "/content/configs/qwen36_27b_qlora_48k_openhands_segments_v2.yaml", "base_model": "Qwen/Qwen3.6-27B", "base_model_config": "Qwen/Qwen3.6-27B", "batch_size": 8, "bf16": true, "capabilities": { "bf16": true, "compute_capability": "sm_80", "fp8": false, "n_gpu": 1, "n_node": 1, "tf32": true }, "context_parallel_size": 1, "cut_cross_entropy": true, "dataloader_num_workers": 2, "dataloader_prefetch_factor": 2, "dataset_num_proc": 12, "dataset_prepared_path": "/content/outputs/last_run_prepared_48k_openhands_segments_v2", "datasets": [ { "message_property_mappings": { "content": "content", "role": "role" }, "path": "/content/data/segments.train.jsonl", "trust_remote_code": false, "type": "input_output" } ], "ddp": false, "device": "cuda:0", "device_map": "auto", "dion_rank_fraction": 1.0, "dion_rank_multiple_of": 1, "eaft_alpha": 1.0, "eaft_k": 20, "embeddings_skip_upcast": true, "env_capabilities": { "torch_version": "2.10.0" }, "eot_tokens": [ "<|im_end|>" ], "eval_batch_size": 1, "eval_causal_lm_metrics": [ "sacrebleu", "comet", "ter", "chrf" ], "eval_max_new_tokens": 128, "eval_sample_packing": true, "eval_steps": 0.5, "eval_table_size": 0, "evals_per_epoch": 1, "experimental_skip_move_to_device": true, "flash_attention": 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 }, "greater_is_better": false, "include_tkps": true, "is_multimodal": true, "layer_offloading": false, "learning_rate": 0.0002, "lisa_layers_attribute": "model.layers", "load_best_model_at_end": true, "load_in_4bit": true, "load_in_8bit": false, "local_rank": 0, "logging_steps": 1, "lora_alpha": 128, "lora_dropout": 0.0, "lora_mlp_kernel": true, "lora_model_dir": "/content/outputs/qwen36-27b-qlora-48k-openhands", "lora_o_kernel": true, "lora_qkv_kernel": true, "lora_r": 64, "lora_target_modules": [ "q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj" ], "loraplus_lr_embedding": 1e-06, "lr_scheduler": "cosine", "mean_resizing_embeddings": false, "merge_method": "memory_efficient", "metric_for_best_model": "eval_loss", "micro_batch_size": 1, "model_config_type": "qwen3_5", "model_config_type_text": "qwen3_5_text", "num_epochs": 2.0, "num_generation_samples": 3, "optimizer": "paged_adamw_8bit", "otel_metrics_host": "localhost", "otel_metrics_port": 8000, "output_dir": "/content/outputs/qwen36-27b-qlora-48k-openhands/merged", "pad_to_sequence_len": false, "plugins": [ "axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin" ], "pretrain_multipack_attn": true, "processor_config": "Qwen/Qwen3.6-27B", "profiler_steps_start": 0, "qlora_sharded_model_loading": false, "quantize_moe_experts": false, "ray_num_workers": 1, "resources_per_worker": { "GPU": 1 }, "sample_packing": true, "sample_packing_bin_size": 200, "sample_packing_group_size": 100000, "save_only_model": false, "save_safetensors": true, "save_steps": 0.25, "save_total_limit": 6, "saves_per_epoch": 2, "seed": 42, "sequence_len": 49152, "shuffle_before_merging_datasets": false, "shuffle_merged_datasets": true, "skip_prepare_dataset": false, "special_tokens": { "eos_token": "<|im_end|>", "pad_token": "<|endoftext|>" }, "streaming_multipack_buffer_size": 10000, "strict": false, "tensor_parallel_size": 1, "test_datasets": [ { "message_property_mappings": { "content": "content", "role": "role" }, "path": "/content/data/segments.val.jsonl", "trust_remote_code": false, "type": "input_output" } ], "tf32": true, "tiled_mlp": true, "tiled_mlp_use_original_mlp": true, "tokenizer_config": "Qwen/Qwen3.6-27B", "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": 0.03, "weight_decay": 0.0, "world_size": 1 } [2026-05-20 01:49:25,321] [INFO] [axolotl.integrations.base.register:380] [PID:100066] Attempting to load plugin: axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin [2026-05-20 01:49:25,321] [INFO] [axolotl.integrations.base.register:383] [PID:100066] Plugin loaded successfully: axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin [2026-05-20 01:49:25,392] [WARNING] [axolotl.utils.schemas.validation.check_sample_packing_without_attention:190] [PID:100066] sample_packing without flash, sdp, xformers, sage, or flex attention does not handle cross sample decontamination. [2026-05-20 01:49:25,392] [WARNING] [axolotl.utils.schemas.validation.hint_sample_packing_padding:235] [PID:100066] `pad_to_sequence_len: true` is recommended when using sample_packing [2026-05-20 01:49:25,540] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:100066] baseline 0.000GB () [2026-05-20 01:49:25,541] [INFO] [axolotl.cli.config.load_cfg:341] [PID:100066] config: { "activation_offloading": false, "adapter": "qlora", "attn_implementation": "flash_attention_2", "axolotl_config_path": "/content/configs/qwen36_27b_qlora_48k_openhands_segments_v2.yaml", "base_model": "Qwen/Qwen3.6-27B", "base_model_config": "Qwen/Qwen3.6-27B", "batch_size": 8, "bf16": true, "capabilities": { "bf16": true, "compute_capability": "sm_80", "fp8": false, "n_gpu": 1, "n_node": 1, "tf32": true }, "context_parallel_size": 1, "cut_cross_entropy": true, "dataloader_num_workers": 2, "dataloader_prefetch_factor": 2, "dataset_num_proc": 12, "dataset_prepared_path": "/content/outputs/last_run_prepared_48k_openhands_segments_v2", "datasets": [ { "message_property_mappings": { "content": "content", "role": "role" }, "path": "/content/data/segments.train.jsonl", "trust_remote_code": false, "type": "input_output" } ], "ddp": false, "device": "cuda:0", "device_map": "auto", "dion_rank_fraction": 1.0, "dion_rank_multiple_of": 1, "eaft_alpha": 1.0, "eaft_k": 20, "embeddings_skip_upcast": true, "env_capabilities": { "torch_version": "2.10.0" }, "eot_tokens": [ "<|im_end|>" ], "eval_batch_size": 1, "eval_causal_lm_metrics": [ "sacrebleu", "comet", "ter", "chrf" ], "eval_max_new_tokens": 128, "eval_sample_packing": true, "eval_steps": 0.5, "eval_table_size": 0, "evals_per_epoch": 1, "experimental_skip_move_to_device": true, "flash_attention": false, "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 }, "greater_is_better": false, "include_tkps": true, "is_multimodal": true, "layer_offloading": false, "learning_rate": 0.0002, "lisa_layers_attribute": "model.layers", "load_best_model_at_end": true, "load_in_4bit": false, "load_in_8bit": false, "local_rank": 0, "logging_steps": 1, "lora_alpha": 128, "lora_dropout": 0.0, "lora_mlp_kernel": true, "lora_model_dir": "/content/outputs/qwen36-27b-qlora-48k-openhands", "lora_o_kernel": true, "lora_qkv_kernel": true, "lora_r": 64, "lora_target_modules": [ "q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj" ], "loraplus_lr_embedding": 1e-06, "lr_scheduler": "cosine", "mean_resizing_embeddings": false, "merge_lora": true, "merge_method": "memory_efficient", "metric_for_best_model": "eval_loss", "micro_batch_size": 1, "model_config_type": "qwen3_5", "model_config_type_text": "qwen3_5_text", "num_epochs": 2.0, "num_generation_samples": 3, "optimizer": "paged_adamw_8bit", "otel_metrics_host": "localhost", "otel_metrics_port": 8000, "output_dir": "/content/outputs/qwen36-27b-qlora-48k-openhands/merged", "pad_to_sequence_len": false, "plugins": [ "axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin" ], "pretrain_multipack_attn": true, "processor_config": "Qwen/Qwen3.6-27B", "profiler_steps_start": 0, "qlora_sharded_model_loading": false, "quantize_moe_experts": false, "ray_num_workers": 1, "resources_per_worker": { "GPU": 1 }, "sample_packing": true, "sample_packing_bin_size": 200, "sample_packing_group_size": 100000, "save_only_model": false, "save_safetensors": true, "save_steps": 0.25, "save_total_limit": 6, "saves_per_epoch": 2, "seed": 42, "sequence_len": 49152, "shuffle_before_merging_datasets": false, "shuffle_merged_datasets": true, "skip_prepare_dataset": false, "special_tokens": { "eos_token": "<|im_end|>", "pad_token": "<|endoftext|>" }, "streaming_multipack_buffer_size": 10000, "strict": false, "tensor_parallel_size": 1, "test_datasets": [ { "message_property_mappings": { "content": "content", "role": "role" }, "path": "/content/data/segments.val.jsonl", "trust_remote_code": false, "type": "input_output" } ], "tf32": true, "tiled_mlp": true, "tiled_mlp_use_original_mlp": true, "tokenizer_config": "Qwen/Qwen3.6-27B", "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": 0.03, "weight_decay": 0.0, "world_size": 1 } [2026-05-20 01:49:25,541] [DEBUG] [axolotl.cli.merge_lora.do_merge_lora:32] [PID:100066] Using memory-efficient LoRA merging method... [2026-05-20 01:49:25,541] [DEBUG] [axolotl.cli.merge_lora._do_merge_lora_efficient:79] [PID:100066] Using memory-efficient LoRA merging method... Fetching 29 files: 0%| | 0/29 [00:00