diff --git "a/debug.log" "b/debug.log" new file mode 100644--- /dev/null +++ "b/debug.log" @@ -0,0 +1,3081 @@ +[2025-11-28 18:02:44,648] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:3633661] baseline 0.000GB () +[2025-11-28 18:02:44,648] [INFO] [axolotl.cli.config.load_cfg:248] [PID:3633661] config: +{ + "activation_offloading": false, + "adapter": "lora", + "axolotl_config_path": "/config.yaml", + "base_model": "mistralai/Mistral-Nemo-Instruct-2407", + "base_model_config": "mistralai/Mistral-Nemo-Instruct-2407", + "batch_size": 9, + "bf16": true, + "bnb_config_kwargs": { + "bnb_4bit_compute_dtype": "bfloat16", + "bnb_4bit_quant_type": "nf4", + "bnb_4bit_use_double_quant": true + }, + "capabilities": { + "bf16": true, + "compute_capability": "sm_90", + "fp8": false, + "n_gpu": 1, + "n_node": 1 + }, + "chat_template": "tokenizer_default", + "context_parallel_size": 1, + "dataloader_num_workers": 1, + "dataloader_pin_memory": true, + "dataloader_prefetch_factor": 256, + "dataset_num_proc": 96, + "datasets": [ + { + "chat_template": "tokenizer_default", + "conversation": "messages", + "message_property_mappings": { + "content": "content", + "role": "role" + }, + "path": "TeamPV/distractors-onr-v2", + "split": "train", + "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": "2.8.0" + }, + "eval_batch_size": 14, + "eval_causal_lm_metrics": [ + "sacrebleu", + "comet", + "ter", + "chrf" + ], + "eval_max_new_tokens": 128, + "eval_sample_packing": false, + "eval_steps": 0.05, + "eval_table_size": 0, + "evals_per_epoch": 5, + "experimental_skip_move_to_device": true, + "flash_attention": true, + "fp16": false, + "gradient_accumulation_steps": 1, + "gradient_checkpointing": true, + "gradient_checkpointing_kwargs": { + "use_reentrant": false + }, + "hf_use_auth_token": true, + "hub_model_id": "TeamPV/mistral-nemo-onr-sft", + "hub_strategy": "every_save", + "include_tkps": true, + "is_mistral_derived_model": true, + "learning_rate": 5e-05, + "lisa_layers_attribute": "model.layers", + "load_best_model_at_end": true, + "load_in_4bit": false, + "load_in_8bit": false, + "local_rank": 0, + "logging_steps": 50, + "lora_alpha": 64, + "lora_dropout": 0.05, + "lora_r": 32, + "lora_target_modules": [ + "q_proj", + "k_proj", + "v_proj", + "o_proj", + "up_proj", + "down_proj", + "gate_proj" + ], + "loraplus_lr_embedding": 1e-06, + "lr_scheduler": "cosine", + "mean_resizing_embeddings": false, + "micro_batch_size": 9, + "model_config_type": "mistral", + "num_epochs": 4.0, + "optimizer": "adamw_bnb_8bit", + "otel_metrics_host": "localhost", + "otel_metrics_port": 8000, + "output_dir": "/model_out/mistral-nemo-12b_sft", + "pad_to_sequence_len": true, + "peft_use_dora": false, + "pretrain_multipack_attn": true, + "profiler_steps_start": 0, + "qlora_sharded_model_loading": false, + "ray_num_workers": 1, + "resources_per_worker": { + "GPU": 1 + }, + "sample_packing": false, + "sample_packing_bin_size": 200, + "sample_packing_group_size": 100000, + "save_first_step": true, + "save_only_model": false, + "save_safetensors": true, + "save_steps": 0.125, + "save_total_limit": 10, + "saves_per_epoch": 2, + "seed": 42, + "sequence_len": 3000, + "shuffle_before_merging_datasets": false, + "shuffle_merged_datasets": true, + "skip_prepare_dataset": false, + "special_tokens": { + "pad_token": "" + }, + "streaming_multipack_buffer_size": 10000, + "strict": false, + "tensor_parallel_size": 1, + "tf32": true, + "tiled_mlp_use_original_mlp": true, + "tokenizer_config": "mistralai/Mistral-Nemo-Instruct-2407", + "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_otel_metrics": false, + "use_ray": false, + "use_tensorboard": true, + "val_set_size": 0.01, + "vllm": { + "device": "auto", + "dtype": "auto", + "gpu_memory_utilization": 0.9, + "host": "0.0.0.0", + "port": 8000 + }, + "warmup_ratio": 0.1, + "weight_decay": 0.0, + "world_size": 1 +} +[2025-11-28 18:02:45,557] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:278] [PID:3633661] EOS: 2 / +[2025-11-28 18:02:45,557] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:3633661] BOS: 1 / +[2025-11-28 18:02:45,557] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:3633661] PAD: 2 / +[2025-11-28 18:02:45,557] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:3633661] UNK: 0 / +[2025-11-28 18:02:45,565] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:481] [PID:3633661] Unable to find prepared dataset in last_run_prepared/0ddf8b5d6caf21513bfb49ac1f784d2b +[2025-11-28 18:02:45,565] [INFO] [axolotl.utils.data.sft._load_raw_datasets:320] [PID:3633661] Loading raw datasets... +[2025-11-28 18:02:45,565] [WARNING] [axolotl.utils.data.sft._load_raw_datasets:322] [PID:3633661] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset using `axolotl preprocess path/to/config.yml`. +[2025-11-28 18:02:46,489] [WARNING] [huggingface_hub.repocard.content:108] [PID:3633661] Repo card metadata block was not found. Setting CardData to empty. +[2025-11-28 18:02:48,633] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:3633661] Loading dataset: TeamPV/distractors-onr-v2 with base_type: chat_template and prompt_style: None +[2025-11-28 18:02:48,642] [INFO] [axolotl.prompt_strategies.chat_template.__call__:969] [PID:3633661] Using chat template: +--- +{%- if messages[0]["role"] == "system" %} + {%- set system_message = messages[0]["content"] %} + {%- set loop_messages = messages[1:] %} +{%- else %} + {%- set loop_messages = messages %} +{%- endif %} +{%- if not tools is defined %} + {%- set tools = none %} +{%- endif %} +{%- set user_messages = loop_messages | selectattr("role", "equalto", "user") | list %} + +{#- This block checks for alternating user/assistant messages, skipping tool calling messages #} +{%- set ns = namespace() %} +{%- set ns.index = 0 %} +{%- for message in loop_messages %} + {%- if not (message.role == "tool" or message.role == "tool_results" or (message.tool_calls is defined and message.tool_calls is not none)) %} + {%- if (message["role"] == "user") != (ns.index % 2 == 0) %} + {{- raise_exception("After the optional system message, conversation roles must alternate user/assistant/user/assistant/...") }} + {%- endif %} + {%- set ns.index = ns.index + 1 %} + {%- endif %} +{%- endfor %} + +{{- bos_token }} +{%- for message in loop_messages %} + {%- if message["role"] == "user" %} + {%- if tools is not none and (message == user_messages[-1]) %} + {{- "[AVAILABLE_TOOLS][" }} + {%- for tool in tools %} + {%- set tool = tool.function %} + {{- '{"type": "function", "function": {' }} + {%- for key, val in tool.items() if key != "return" %} + {%- if val is string %} + {{- '"' + key + '": "' + val + '"' }} + {%- else %} + {{- '"' + key + '": ' + val|tojson }} + {%- endif %} + {%- if not loop.last %} + {{- ", " }} + {%- endif %} + {%- endfor %} + {{- "}}" }} + {%- if not loop.last %} + {{- ", " }} + {%- else %} + {{- "]" }} + {%- endif %} + {%- endfor %} + {{- "[/AVAILABLE_TOOLS]" }} + {%- endif %} + {%- if loop.last and system_message is defined %} + {{- "[INST]" + system_message + "\n\n" + message["content"] + "[/INST]" }} + {%- else %} + {{- "[INST]" + message["content"] + "[/INST]" }} + {%- endif %} + {%- elif (message.tool_calls is defined and message.tool_calls is not none) %} + {{- "[TOOL_CALLS][" }} + {%- for tool_call in message.tool_calls %} + {%- set out = tool_call.function|tojson %} + {{- out[:-1] }} + {%- if not tool_call.id is defined or tool_call.id|length != 9 %} + {{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }} + {%- endif %} + {{- ', "id": "' + tool_call.id + '"}' }} + {%- if not loop.last %} + {{- ", " }} + {%- else %} + {{- "]" + eos_token }} + {%- endif %} + {%- endfor %} + {%- elif message["role"] == "assistant" %} + {{- message["content"] + eos_token}} + {%- elif message["role"] == "tool_results" or message["role"] == "tool" %} + {%- if message.content is defined and message.content.content is defined %} + {%- set content = message.content.content %} + {%- else %} + {%- set content = message.content %} + {%- endif %} + {{- '[TOOL_RESULTS]{"content": ' + content|string + ", " }} + {%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %} + {{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }} + {%- endif %} + {{- '"call_id": "' + message.tool_call_id + '"}[/TOOL_RESULTS]' }} + {%- else %} + {{- raise_exception("Only user and assistant roles are supported, with the exception of an initial optional system message!") }} + {%- endif %} +{%- endfor %} + +--- +[2025-11-28 18:03:01,375] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:218] [PID:3633661] min_input_len: 168 +[2025-11-28 18:03:01,375] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:220] [PID:3633661] max_input_len: 4582 +[2025-11-28 18:03:03,407] [WARNING] [axolotl.utils.data.utils.handle_long_seq_in_dataset:260] [PID:3633661] Dropped 204 samples from dataset + Saving the dataset (0/96 shards): 0%| | 0/99861 [00:00 +[2025-11-28 18:03:19,445] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:3633661] BOS: 1 / +[2025-11-28 18:03:19,445] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:3633661] PAD: 2 / +[2025-11-28 18:03:19,445] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:3633661] UNK: 0 / +[2025-11-28 18:03:19,445] [DEBUG] [axolotl.train.setup_model_and_tokenizer:74] [PID:3633661] Loading model +[2025-11-28 18:03:19,578] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:87] [PID:3633661] Patched Trainer.evaluation_loop with nanmean loss calculation +[2025-11-28 18:03:19,579] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:138] [PID:3633661] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation + Loading checkpoint shards: 0%| | 0/5 [00:00