[2026-04-09 19:06:27,798] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:4429] baseline 0.000GB () [2026-04-09 19:06:27,800] [INFO] [axolotl.cli.config.load_cfg:248] [PID:4429] config: { "activation_offloading": false, "adapter": "lora", "axolotl_config_path": "config.yaml", "base_model": "meta-llama/Llama-3.1-8B-Instruct", "base_model_config": "meta-llama/Llama-3.1-8B-Instruct", "batch_size": 8, "bf16": true, "capabilities": { "bf16": true, "compute_capability": "sm_90", "fp8": false, "n_gpu": 1, "n_node": 1 }, "context_parallel_size": 1, "dataloader_num_workers": 1, "dataloader_pin_memory": true, "dataloader_prefetch_factor": 256, "dataset_processes": 26, "datasets": [ { "chat_template": "llama3", "field_messages": "messages", "message_property_mappings": { "content": "content", "role": "role" }, "path": "mx003/cve", "trust_remote_code": false, "type": "chat_template" } ], "ddp": false, "device": "cuda:0", "dion_rank_fraction": 1.0, "dion_rank_multiple_of": 1, "env_capabilities": { "torch_version": "2.7.1" }, "eval_batch_size": 2, "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, "gradient_accumulation_steps": 4, "gradient_checkpointing": true, "gradient_checkpointing_kwargs": { "use_reentrant": true }, "group_by_length": true, "include_tkps": true, "is_llama_derived_model": true, "learning_rate": 0.0002, "lisa_layers_attribute": "model.layers", "load_best_model_at_end": false, "load_in_4bit": false, "load_in_8bit": false, "local_rank": 0, "lora_alpha": 64, "lora_dropout": 0.05, "lora_r": 32, "lora_target_modules": [ "q_proj", "v_proj", "k_proj", "o_proj", "gate_proj", "down_proj", "up_proj" ], "loraplus_lr_embedding": 1e-06, "lr_scheduler": "cosine", "mean_resizing_embeddings": false, "micro_batch_size": 2, "model_config_type": "llama", "num_epochs": 3.0, "optimizer": "adamw_torch_fused", "output_dir": "./outputs/mymodel", "pad_to_sequence_len": true, "pretrain_multipack_attn": true, "profiler_steps_start": 0, "qlora_sharded_model_loading": 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": 50, "sequence_len": 4096, "shuffle_before_merging_datasets": false, "shuffle_merged_datasets": true, "skip_prepare_dataset": false, "special_tokens": { "pad_token": "<|end_of_text|>" }, "streaming_multipack_buffer_size": 10000, "strict": false, "tensor_parallel_size": 1, "tiled_mlp_use_original_mlp": true, "tokenizer_config": "meta-llama/Llama-3.1-8B-Instruct", "tokenizer_save_jinja_files": true, "torch_dtype": "torch.bfloat16", "train_on_inputs": false, "trl": { "log_completions": false, "mask_truncated_completions": false, "ref_model_mixup_alpha": 0.9, "ref_model_sync_steps": 64, "scale_rewards": true, "sync_ref_model": false, "use_vllm": false, "vllm_server_host": "0.0.0.0", "vllm_server_port": 8000 }, "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 }, "weight_decay": 0.0, "world_size": 1 } [2026-04-09 19:06:28,524] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:278] [PID:4429] EOS: 128009 / <|eot_id|> [2026-04-09 19:06:28,525] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:4429] BOS: 128000 / <|begin_of_text|> [2026-04-09 19:06:28,526] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:4429] PAD: 128001 / <|end_of_text|> [2026-04-09 19:06:28,527] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:4429] UNK: None / None [2026-04-09 19:06:28,529] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:476] [PID:4429] Unable to find prepared dataset in last_run_prepared/cf56a7ab70db915ab24addbfc4ad78b9 [2026-04-09 19:06:28,530] [INFO] [axolotl.utils.data.sft._load_raw_datasets:320] [PID:4429] Loading raw datasets... [2026-04-09 19:06:28,531] [WARNING] [axolotl.utils.data.sft._load_raw_datasets:322] [PID:4429] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset using `axolotl preprocess path/to/config.yml`. [2026-04-09 19:06:30,317] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:4429] Loading dataset: mx003/cve with base_type: chat_template and prompt_style: None [2026-04-09 19:06:30,323] [INFO] [axolotl.prompt_strategies.chat_template.__call__:969] [PID:4429] Using chat template: --- {% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|> '+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|> ' }}{% endif %} --- Tokenizing Prompts (num_proc=26): 0%| | 0/485 [00:004096) (num_proc=26): 0%| | 0/485 [00:004096) (num_proc=26): 4%|██▍ | 19/485 [00:00<00:11, 42.21 examples/s] Dropping Long Sequences (>4096) (num_proc=26): 39%|███████████████████████▌ | 190/485 [00:00<00:00, 438.95 examples/s] Dropping Long Sequences (>4096) (num_proc=26): 89%|█████████████████████████████████████████████████████▎ | 431/485 [00:00<00:00, 928.32 examples/s] Dropping Long Sequences (>4096) (num_proc=26): 100%|████████████████████████████████████████████████████████████| 485/485 [00:00<00:00, 555.88 examples/s] [2026-04-09 19:06:36,107] [WARNING] [axolotl.utils.data.utils.handle_long_seq_in_dataset:260] [PID:4429] Dropped 26 samples from dataset Drop Samples with Zero Trainable Tokens (num_proc=26): 0%| | 0/459 [00:00 [2026-04-09 19:06:42,810] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:4429] BOS: 128000 / <|begin_of_text|> [2026-04-09 19:06:42,811] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:4429] PAD: 128001 / <|end_of_text|> [2026-04-09 19:06:42,812] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:4429] UNK: None / None [2026-04-09 19:06:42,813] [DEBUG] [axolotl.train.setup_model_and_tokenizer:74] [PID:4429] Loading model [2026-04-09 19:06:42,927] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:87] [PID:4429] Patched Trainer.evaluation_loop with nanmean loss calculation [2026-04-09 19:06:42,929] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:138] [PID:4429] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation [2026-04-09 19:06:42,931] [INFO] [axolotl.loaders.patch_manager._apply_multipack_patches:301] [PID:4429] Applying multipack dataloader patch for sample packing... Loading checkpoint shards: 0%| | 0/4 [00:00