[2026-03-21 00:18:14,920] [DEBUG] [axolotl.utils.config.resolve_dtype:69] [PID:3143] bf16 support detected, enabling for this configuration. [2026-03-21 00:18:15,065] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:3143] baseline 0.000GB () [2026-03-21 00:18:15,066] [INFO] [axolotl.cli.config.load_cfg:341] [PID:3143] config: { "activation_offloading": true, "adapter": "lora", "axolotl_config_path": "train.yml", "base_model": "Qwen/Qwen3.5-27B", "base_model_config": "Qwen/Qwen3.5-27B", "batch_size": 32, "bf16": true, "capabilities": { "bf16": true, "compute_capability": "sm_90", "fp8": true, "n_gpu": 1, "n_node": 1, "tf32": true }, "chat_template": "tokenizer_default", "context_parallel_size": 1, "cosine_min_lr_ratio": 0.1, "cut_cross_entropy": true, "dataloader_num_workers": 1, "dataloader_pin_memory": true, "dataloader_prefetch_factor": 256, "dataset_num_proc": 24, "dataset_prepared_path": "last_run_prepared", "datasets": [ { "ds_type": "parquet", "message_property_mappings": { "content": "content", "role": "role" }, "path": "output.parquet", "trust_remote_code": false } ], "ddp": false, "device": "cuda:0", "dion_rank_fraction": 1.0, "dion_rank_multiple_of": 1, "eaft_alpha": 1.0, "eaft_k": 20, "env_capabilities": { "torch_version": "2.8.0" }, "eval_batch_size": 8, "eval_causal_lm_metrics": [ "sacrebleu", "comet", "ter", "chrf" ], "eval_max_new_tokens": 128, "eval_sample_packing": true, "eval_table_size": 0, "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": 4, "gradient_checkpointing": true, "gradient_checkpointing_kwargs": { "use_reentrant": true }, "group_by_length": false, "include_tkps": true, "is_multimodal": true, "learning_rate": 1e-05, "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": 512, "lora_dropout": 0.0, "lora_r": 64, "lora_target_modules": [ "q_proj", "k_proj", "v_proj", "o_proj", "down_proj", "up_proj", "linear_attn.in_proj_qkv", "linear_attn.in_proj_z", "linear_attn.out_proj" ], "loraplus_lr_embedding": 1e-06, "lr_scheduler": "constant", "max_grad_norm": 0.1, "mean_resizing_embeddings": false, "micro_batch_size": 8, "model_config_type": "qwen3_5", "model_config_type_text": "qwen3_5_text", "num_epochs": 2.0, "num_generation_samples": 3, "optimizer": "adamw_torch_8bit", "otel_metrics_host": "localhost", "otel_metrics_port": 8000, "output_dir": "./model-output", "pad_to_sequence_len": true, "plugins": [ "axolotl.integrations.liger.LigerPlugin", "axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin" ], "pretrain_multipack_attn": true, "processor_config": "Qwen/Qwen3.5-27B", "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": 0.5, "saves_per_epoch": 1, "sequence_len": 8192, "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": "Qwen/Qwen3.5-27B", "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, "use_wandb": true, "val_set_size": 0.0, "vllm": { "device": "auto", "dtype": "auto", "gpu_memory_utilization": 0.9, "host": "0.0.0.0", "port": 8000 }, "wandb_project": "qwen-27b-seemo", "warmup_ratio": 0.05, "weight_decay": 0.001, "world_size": 1 } [2026-03-21 00:18:15,069] [WARNING] [axolotl.cli.checks.check_user_token:46] [PID:3143] Error verifying HuggingFace token. Remember to log in using `hf auth login` and get your access token from https://huggingface.co/settings/tokens if you want to use gated models or datasets. [2026-03-21 00:18:15,157] [WARNING] [huggingface_hub.utils._http._warn_on_warning_headers:916] [PID:3143] Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads. [2026-03-21 00:18:16,548] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:299] [PID:3143] EOS: 248046 / <|im_end|> [2026-03-21 00:18:16,548] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:300] [PID:3143] BOS: None / None [2026-03-21 00:18:16,548] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:301] [PID:3143] PAD: 248044 / <|endoftext|> [2026-03-21 00:18:16,548] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:302] [PID:3143] UNK: None / None [2026-03-21 00:18:16,550] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:475] [PID:3143] Loading prepared dataset from disk at last_run_prepared/e72d0f3749caf5df6eb0a14757b8955e... Loading dataset from disk: 0%| | 0/24 [00:00 [2026-03-21 00:18:22,269] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:300] [PID:3143] BOS: None / None [2026-03-21 00:18:22,269] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:301] [PID:3143] PAD: 248044 / <|endoftext|> [2026-03-21 00:18:22,269] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:302] [PID:3143] UNK: None / None preprocessor_config.json: 0%| | 0.00/390 [00:00