Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{%- for message in messages -%}{%- if loop.first and message['role'] != 'system' -%}{{ '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}{%- endif -%}{{ '<|im_start|>' + message['role'] + '\\n' }}{%- if message['content'] is string -%}{{- message['content'] -}}{%- elif message['content'] is iterable -%}{%- for item in message['content'] -%}{%- if item['type'].startswith('image') -%}{%- set image_count.value = image_count.value + 1 -%}{%- if add_vision_id -%}{{ 'Picture ' + image_count.value + ': ' }}{%- endif -%}{{ '<|vision_start|><|image_pad|><|vision_end|>' }}{%- elif item['type'].startswith('video') -%}{%- set video_count.value = video_count.value + 1 -%}{%- if add_vision_id -%}{{ 'Video ' + video_count.value + ': ' }}{%- endif -%}{{ '<|vision_start|><|video_pad|><|vision_end|>' }}{%- elif item['type']=='text' -%}{{- item['text'] if 'text' in item else item['content'] -}}{%- endif -%}{%- endfor -%}{%- endif -%}{{ '<|im_end|>\\n' }}{%- endfor -%}{%- if add_generation_prompt -%}{{- '<|im_start|>assistant\\n' -}}{%- endif -%}"
+}

config.json

@@ -0,0 +1,65 @@
+{
+  "architectures": [
+    "Qwen2_5_VLForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "image_token_id": 151655,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 128000,
+  "max_window_layers": 28,
+  "model_type": "qwen2_5_vl",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": {
+    "mrope_section": [
+      16,
+      24,
+      24
+    ],
+    "rope_type": "default",
+    "type": "default"
+  },
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "depth": 32,
+    "fullatt_block_indexes": [
+      7,
+      15,
+      23,
+      31
+    ],
+    "hidden_act": "silu",
+    "hidden_size": 1280,
+    "in_channels": 3,
+    "in_chans": 3,
+    "intermediate_size": 3420,
+    "model_type": "qwen2_5_vl",
+    "num_heads": 16,
+    "out_hidden_size": 3584,
+    "patch_size": 14,
+    "spatial_merge_size": 2,
+    "spatial_patch_size": 14,
+    "temporal_patch_size": 2,
+    "tokens_per_second": 2,
+    "torch_dtype": "bfloat16",
+    "window_size": 112
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 152064
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 1e-06,
+  "transformers_version": "4.51.3"
+}

logs/rank_0000.log

@@ -0,0 +1,396 @@
+[2025-07-04 10:07:00,006][oumi][rank0][pid:2619050][MainThread][INFO]][train.py:283] TrainingConfig:
+TrainingConfig(data=DataParams(train=DatasetSplitParams(datasets=[DatasetParams(dataset_name='hf_vision',
+                                                                                dataset_path=None,
+                                                                                subset=None,
+                                                                                split='train',
+                                                                                dataset_kwargs={'answer_column': 'solution',
+                                                                                                'hf_dataset_path': 'penfever/MM-MathInstruct-to-r1-format-filtered',
+                                                                                                'image_column': 'image',
+                                                                                                'processor_name': 'Qwen/Qwen2.5-VL-7B-Instruct',
+                                                                                                'question_column': 'problem',
+                                                                                                'return_conversations': True,
+                                                                                                'return_tensors': True},
+                                                                                sample_count=None,
+                                                                                mixture_proportion=None,
+                                                                                shuffle=True,
+                                                                                seed=42,
+                                                                                shuffle_buffer_size=1000,
+                                                                                trust_remote_code=True,
+                                                                                transform_num_workers='auto')],
+                                                        collator_name='vision_language_sft',
+                                                        collator_kwargs={'process_individually': True},
+                                                        pack=False,
+                                                        stream=False,
+                                                        target_col=None,
+                                                        mixture_strategy='first_exhausted',
+                                                        seed=None,
+                                                        use_async_dataset=False,
+                                                        use_torchdata=True),
+                               test=DatasetSplitParams(datasets=[],
+                                                       collator_name=None,
+                                                       collator_kwargs={},
+                                                       pack=False,
+                                                       stream=False,
+                                                       target_col=None,
+                                                       mixture_strategy='first_exhausted',
+                                                       seed=None,
+                                                       use_async_dataset=False,
+                                                       use_torchdata=None),
+                               validation=DatasetSplitParams(datasets=[],
+                                                             collator_name=None,
+                                                             collator_kwargs={},
+                                                             pack=False,
+                                                             stream=False,
+                                                             target_col=None,
+                                                             mixture_strategy='first_exhausted',
+                                                             seed=None,
+                                                             use_async_dataset=False,
+                                                             use_torchdata=None)),
+               model=ModelParams(model_name='Qwen/Qwen2.5-VL-7B-Instruct',
+                                 adapter_model=None,
+                                 tokenizer_name=None,
+                                 tokenizer_pad_token=None,
+                                 tokenizer_kwargs={},
+                                 processor_kwargs={},
+                                 model_max_length=10000,
+                                 load_pretrained_weights=True,
+                                 trust_remote_code=True,
+                                 torch_dtype_str='bfloat16',
+                                 compile=False,
+                                 chat_template='qwen2-vl-instruct',
+                                 attn_implementation='sdpa',
+                                 device_map='auto',
+                                 model_kwargs={},
+                                 enable_liger_kernel=False,
+                                 shard_for_eval=False,
+                                 freeze_layers=[],
+                                 model_revision=None),
+               training=TrainingParams(use_peft=False,
+                                       trainer_type=<TrainerType.TRL_SFT: 'trl_sft'>,
+                                       enable_gradient_checkpointing=True,
+                                       gradient_checkpointing_kwargs={'use_reentrant': False},
+                                       output_dir='output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered',
+                                       per_device_train_batch_size=1,
+                                       per_device_eval_batch_size=8,
+                                       gradient_accumulation_steps=1,
+                                       max_steps=3750,
+                                       num_train_epochs=5,
+                                       save_epoch=False,
+                                       save_steps=0,
+                                       save_final_model=True,
+                                       seed=42,
+                                       data_seed=42,
+                                       use_deterministic=False,
+                                       full_determinism=False,
+                                       run_name=None,
+                                       metrics_function=None,
+                                       reward_functions=None,
+                                       grpo=GrpoParams(model_init_kwargs={},
+                                                       max_prompt_length=None,
+                                                       max_completion_length=None,
+                                                       num_generations=None,
+                                                       temperature=0.9,
+                                                       remove_unused_columns=False,
+                                                       repetition_penalty=1.0,
+                                                       use_vllm=False,
+                                                       vllm_device=None,
+                                                       vllm_gpu_memory_utilization=0.9,
+                                                       vllm_dtype=None,
+                                                       vllm_max_model_len=None,
+                                                       epsilon=0.2,
+                                                       log_completions=False),
+                                       log_level='info',
+                                       dep_log_level='warning',
+                                       enable_wandb=True,
+                                       enable_mlflow=False,
+                                       enable_tensorboard=True,
+                                       logging_strategy='steps',
+                                       logging_dir=None,
+                                       logging_steps=50,
+                                       logging_first_step=False,
+                                       eval_strategy='no',
+                                       eval_steps=500,
+                                       learning_rate=2e-05,
+                                       lr_scheduler_type='cosine',
+                                       lr_scheduler_kwargs={},
+                                       warmup_ratio=0.03,
+                                       warmup_steps=None,
+                                       optimizer='adamw_torch_fused',
+                                       weight_decay=0.01,
+                                       adam_beta1=0.9,
+                                       adam_beta2=0.999,
+                                       adam_epsilon=1e-08,
+                                       sgd_momentum=0.0,
+                                       mixed_precision_dtype=<MixedPrecisionDtype.NONE: 'none'>,
+                                       compile=False,
+                                       include_performance_metrics=True,
+                                       include_alternative_mfu_metrics=False,
+                                       log_model_summary=False,
+                                       resume_from_checkpoint=None,
+                                       try_resume_from_last_checkpoint=False,
+                                       dataloader_num_workers=2,
+                                       dataloader_persistent_workers=False,
+                                       dataloader_prefetch_factor=8,
+                                       dataloader_main_process_only=False,
+                                       ddp_find_unused_parameters=False,
+                                       max_grad_norm=1.0,
+                                       trainer_kwargs={'dataset_kwargs': {'skip_prepare_dataset': True},
+                                                       'max_seq_length': 10000,
+                                                       'remove_unused_columns': False},
+                                       verl_config_overrides={},
+                                       profiler=ProfilerParams(save_dir=None,
+                                                               enable_cpu_profiling=False,
+                                                               enable_cuda_profiling=False,
+                                                               record_shapes=False,
+                                                               profile_memory=False,
+                                                               with_stack=False,
+                                                               with_flops=False,
+                                                               with_modules=False,
+                                                               row_limit=50,
+                                                               schedule=ProfilerScheduleParams(enable_schedule=False,
+                                                                                               wait=0,
+                                                                                               warmup=1,
+                                                                                               active=3,
+                                                                                               repeat=1,
+                                                                                               skip_first=1)),
+                                       telemetry=TelemetryParams(telemetry_dir='telemetry',
+                                                                 collect_telemetry_for_all_ranks=False,
+                                                                 track_gpu_temperature=False),
+                                       empty_device_cache_steps=1,
+                                       nccl_default_timeout_minutes=None,
+                                       label_ignore_index=None),
+               peft=PeftParams(lora_r=8,
+                               lora_alpha=8,
+                               lora_dropout=0.0,
+                               lora_target_modules=None,
+                               lora_modules_to_save=None,
+                               lora_bias='none',
+                               lora_init_weights=<LoraWeightInitialization.DEFAULT: 'default'>,
+                               lora_task_type=<TaskType.CAUSAL_LM: 'CAUSAL_LM'>,
+                               q_lora=False,
+                               q_lora_bits=4,
+                               bnb_4bit_quant_type='fp4',
+                               llm_int8_skip_modules=None,
+                               use_bnb_nested_quant=False,
+                               bnb_4bit_quant_storage='uint8',
+                               bnb_4bit_compute_dtype='float32',
+                               peft_save_mode=<PeftSaveMode.ADAPTER_ONLY: 'adapter_only'>),
+               fsdp=FSDPParams(enable_fsdp=True,
+                               sharding_strategy=<ShardingStrategy.HYBRID_SHARD: 'HYBRID_SHARD'>,
+                               cpu_offload=False,
+                               mixed_precision='bf16',
+                               backward_prefetch=<BackwardPrefetch.BACKWARD_PRE: 'BACKWARD_PRE'>,
+                               forward_prefetch=True,
+                               use_orig_params=None,
+                               state_dict_type=<StateDictType.FULL_STATE_DICT: 'FULL_STATE_DICT'>,
+                               auto_wrap_policy=<AutoWrapPolicy.SIZE_BASED_WRAP: 'SIZE_BASED_WRAP'>,
+                               min_num_params=100000,
+                               transformer_layer_cls=None,
+                               sync_module_states=True))
+[2025-07-04 10:07:00,227][oumi][rank0][pid:2619050][MainThread][INFO]][models.py:433] Setting tokenizer to use the 'right' padding side for model 'Qwen/Qwen2.5-VL-7B-Instruct'. The 'right' padding side is configured as the default value for this model type.
+[2025-07-04 10:07:02,023][oumi][rank0][pid:2619050][MainThread][INFO]][models.py:479] Using the chat template 'qwen2-vl-instruct' specified in model config for model 'Qwen/Qwen2.5-VL-7B-Instruct'. 
+[2025-07-04 10:07:03,883][oumi][rank0][pid:2619050][MainThread][WARNING]][data.py:66] Using torchdata preprocessing pipeline. This is currently in beta and may not be stable.
+[2025-07-04 10:07:03,883][oumi][rank0][pid:2619050][MainThread][INFO]][base_map_dataset.py:91] Creating map dataset (type: HuggingFaceVisionDataset)... dataset_name: 'penfever/MM-MathInstruct-to-r1-format-filtered'
+[2025-07-04 10:07:05,192][oumi][rank0][pid:2619050][MainThread][INFO]][base_map_dataset.py:487] Dataset Info:
+	Split: train
+	Version: 0.0.0
+	Dataset size: 57058499
+	Download size: 48789762
+	Size: 105848261 bytes
+	Rows: 4286
+	Columns: ['problem', 'solution', 'original_question', 'original_answer', 'image']
+[2025-07-04 10:07:07,950][oumi][rank0][pid:2619050][MainThread][INFO]][base_map_dataset.py:426] Loaded DataFrame with shape: (4286, 5). Columns:
+problem              object
+solution             object
+original_question    object
+original_answer      object
+image                object
+dtype: object
+[2025-07-04 10:07:21,068][oumi][rank0][pid:2619050][MainThread][INFO]][distributed.py:303] Initialized distributed (True): DeviceRankInfo(world_size=8, rank=0, local_world_size=8, local_rank=0)
+[2025-07-04 10:07:21,069][oumi][rank0][pid:2619050][MainThread][INFO]][train.py:406] Set Accelerate environment variables for FSDP: {'ACCELERATE_DYNAMO_BACKEND': 'NO', 'ACCELERATE_DYNAMO_MODE': 'default', 'ACCELERATE_DYNAMO_USE_FULLGRAPH': 'False', 'ACCELERATE_DYNAMO_USE_DYNAMIC': 'False', 'FSDP_CPU_RAM_EFFICIENT_LOADING': 'true', 'FSDP_USE_ORIG_PARAMS': 'true', 'ACCELERATE_USE_FSDP': 'true', 'FSDP_SHARDING_STRATEGY': 'HYBRID_SHARD', 'FSDP_OFFLOAD_PARAMS': 'false', 'ACCELERATE_MIXED_PRECISION': 'bf16', 'FSDP_BACKWARD_PREFETCH': 'BACKWARD_PRE', 'FSDP_FORWARD_PREFETCH': 'true', 'FSDP_STATE_DICT_TYPE': 'FULL_STATE_DICT', 'FSDP_AUTO_WRAP_POLICY': 'SIZE_BASED_WRAP', 'FSDP_MIN_NUM_PARAMS': '100000', 'FSDP_SYNC_MODULE_STATES': 'true', 'FSDP_ACTIVATION_CHECKPOINTING': 'true'}
+[2025-07-04 10:07:21,069][oumi][rank0][pid:2619050][MainThread][INFO]][models.py:219] Accelerate FSDP run detected! Setting device_map to None.
+[2025-07-04 10:07:21,069][oumi][rank0][pid:2619050][MainThread][INFO]][models.py:228] Building model using device_map: None (DeviceRankInfo(world_size=8, rank=0, local_world_size=8, local_rank=0))...
+[2025-07-04 10:07:21,105][oumi][rank0][pid:2619050][MainThread][INFO]][models.py:300] Using model class: <class 'transformers.models.auto.modeling_auto.AutoModelForVision2Seq'> to instantiate model.
+[2025-07-04 10:07:23,607][oumi][rank0][pid:2619050][MainThread][INFO]][torch_utils.py:289] 
+Model Parameters Summary:
+🔢 Total     parameters: 8,292,166,656
+🔗 Embedding parameters: 544,997,376
+🎯 Trainable parameters: 8,292,166,656
+🔒 Frozen    parameters: 0 (0.00%)
+
+[2025-07-04 10:07:24,853][oumi][rank0][pid:2619050][MainThread][INFO]][torch_profiler_utils.py:164] PROF: Torch Profiler disabled!
+[2025-07-04 10:07:24,853][oumi][rank0][pid:2619050][MainThread][WARNING]][callbacks.py:72] MFU logging requires packed datasets. Skipping MFU callbacks.
+[2025-07-04 10:07:25,176][oumi][rank0][pid:2619050][MainThread][INFO]][training.py:62] SFTConfig(output_dir='output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered',
+          overwrite_output_dir=False,
+          do_train=False,
+          do_eval=False,
+          do_predict=False,
+          eval_strategy=<IntervalStrategy.NO: 'no'>,
+          prediction_loss_only=False,
+          per_device_train_batch_size=1,
+          per_device_eval_batch_size=8,
+          per_gpu_train_batch_size=None,
+          per_gpu_eval_batch_size=None,
+          gradient_accumulation_steps=1,
+          eval_accumulation_steps=None,
+          eval_delay=0,
+          torch_empty_cache_steps=1,
+          learning_rate=2e-05,
+          weight_decay=0.01,
+          adam_beta1=0.9,
+          adam_beta2=0.999,
+          adam_epsilon=1e-08,
+          max_grad_norm=1.0,
+          num_train_epochs=5,
+          max_steps=3750,
+          lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>,
+          lr_scheduler_kwargs={},
+          warmup_ratio=0.03,
+          warmup_steps=0,
+          log_level='warning',
+          log_level_replica='warning',
+          log_on_each_node=True,
+          logging_dir='output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered/runs/Jul04_10-07-24_oumi-compute004',
+          logging_strategy=<IntervalStrategy.STEPS: 'steps'>,
+          logging_first_step=False,
+          logging_steps=50,
+          logging_nan_inf_filter=True,
+          save_strategy=<SaveStrategy.NO: 'no'>,
+          save_steps=0,
+          save_total_limit=None,
+          save_safetensors=True,
+          save_on_each_node=False,
+          save_only_model=False,
+          restore_callback_states_from_checkpoint=False,
+          no_cuda=False,
+          use_cpu=False,
+          use_mps_device=False,
+          seed=42,
+          data_seed=42,
+          jit_mode_eval=False,
+          use_ipex=False,
+          bf16=False,
+          fp16=False,
+          fp16_opt_level='O1',
+          half_precision_backend='auto',
+          bf16_full_eval=False,
+          fp16_full_eval=False,
+          tf32=None,
+          local_rank=0,
+          ddp_backend=None,
+          tpu_num_cores=None,
+          tpu_metrics_debug=False,
+          debug=[],
+          dataloader_drop_last=False,
+          eval_steps=500,
+          dataloader_num_workers=2,
+          dataloader_prefetch_factor=8,
+          past_index=-1,
+          run_name='output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered',
+          disable_tqdm=False,
+          remove_unused_columns=False,
+          label_names=None,
+          load_best_model_at_end=False,
+          metric_for_best_model=None,
+          greater_is_better=None,
+          ignore_data_skip=False,
+          fsdp=[],
+          fsdp_min_num_params=0,
+          fsdp_config={'min_num_params': 0,
+                       'xla': False,
+                       'xla_fsdp_grad_ckpt': False,
+                       'xla_fsdp_v2': False},
+          tp_size=0,
+          fsdp_transformer_layer_cls_to_wrap=None,
+          accelerator_config=AcceleratorConfig(split_batches=False,
+                                               dispatch_batches=False,
+                                               even_batches=True,
+                                               use_seedable_sampler=True,
+                                               non_blocking=False,
+                                               gradient_accumulation_kwargs=None,
+                                               use_configured_state=False),
+          deepspeed=None,
+          label_smoothing_factor=0.0,
+          optim=<OptimizerNames.ADAMW_TORCH_FUSED: 'adamw_torch_fused'>,
+          optim_args=None,
+          adafactor=False,
+          group_by_length=False,
+          length_column_name='length',
+          report_to=['wandb', 'tensorboard'],
+          ddp_find_unused_parameters=False,
+          ddp_bucket_cap_mb=None,
+          ddp_broadcast_buffers=None,
+          dataloader_pin_memory=True,
+          dataloader_persistent_workers=False,
+          skip_memory_metrics=True,
+          use_legacy_prediction_loop=False,
+          push_to_hub=False,
+          resume_from_checkpoint=None,
+          hub_model_id=None,
+          hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>,
+          hub_token=None,
+          hub_private_repo=None,
+          hub_always_push=False,
+          gradient_checkpointing=False,
+          gradient_checkpointing_kwargs={'use_reentrant': False},
+          include_inputs_for_metrics=False,
+          include_for_metrics=[],
+          eval_do_concat_batches=True,
+          fp16_backend='auto',
+          push_to_hub_model_id=None,
+          push_to_hub_organization=None,
+          push_to_hub_token=None,
+          mp_parameters='',
+          auto_find_batch_size=False,
+          full_determinism=False,
+          torchdynamo=None,
+          ray_scope='last',
+          ddp_timeout=1800,
+          torch_compile=False,
+          torch_compile_backend=None,
+          torch_compile_mode=None,
+          include_tokens_per_second=True,
+          include_num_input_tokens_seen=True,
+          neftune_noise_alpha=None,
+          optim_target_modules=None,
+          batch_eval_metrics=False,
+          eval_on_start=False,
+          use_liger_kernel=False,
+          eval_use_gather_object=False,
+          average_tokens_across_devices=False,
+          model_init_kwargs=None,
+          dataset_text_field='text',
+          dataset_kwargs={'skip_prepare_dataset': True},
+          dataset_num_proc=None,
+          pad_token=None,
+          max_length=10000,
+          packing=False,
+          padding_free=False,
+          eval_packing=None,
+          dataset_batch_size=None,
+          num_of_sequences=None,
+          chars_per_token=None,
+          max_seq_length=10000,
+          use_liger=None)
+[2025-07-04 10:07:25,232][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] GPU Metrics Before Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=6166.0, temperature=30, fan_speed=None, fan_speeds=None, power_usage_watts=121.434, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 10:07:25,287][oumi][rank0][pid:2619050][MainThread][INFO]][train.py:510] Training init time: 29.326s
+[2025-07-04 10:07:25,287][oumi][rank0][pid:2619050][MainThread][INFO]][train.py:511] Starting training... (TrainerType.TRL_SFT, transformers: 4.51.3)
+[2025-07-04 12:47:15,514][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] On epoch end: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34750.0, temperature=34, fan_speed=None, fan_speeds=None, power_usage_watts=193.085, power_limit_watts=700.0, gpu_utilization=11, memory_utilization=2, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 15:25:44,400][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] On epoch end: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34750.0, temperature=34, fan_speed=None, fan_speeds=None, power_usage_watts=187.86, power_limit_watts=700.0, gpu_utilization=32, memory_utilization=9, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 18:04:39,027][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] On epoch end: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34750.0, temperature=34, fan_speed=None, fan_speeds=None, power_usage_watts=188.725, power_limit_watts=700.0, gpu_utilization=49, memory_utilization=3, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 20:43:20,054][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] On epoch end: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34750.0, temperature=34, fan_speed=None, fan_speeds=None, power_usage_watts=193.78, power_limit_watts=700.0, gpu_utilization=35, memory_utilization=2, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 23:21:14,672][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] On epoch end: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34750.0, temperature=34, fan_speed=None, fan_speeds=None, power_usage_watts=191.45000000000002, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 01:59:05,749][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] On epoch end: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34750.0, temperature=35, fan_speed=None, fan_speeds=None, power_usage_watts=187.851, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:20,509][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] On epoch end: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=34, fan_speed=None, fan_speeds=None, power_usage_watts=191.013, power_limit_watts=700.0, gpu_utilization=36, memory_utilization=3, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:20,705][oumi][rank0][pid:2619050][MainThread][INFO]][telemetry_callback.py:242] Saving telemetry callback summary to output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered/telemetry/telemetry_callback_rank0000.json...
+[2025-07-05 04:37:21,418][oumi][rank0][pid:2619050][MainThread][INFO]][train.py:518] Training is Complete.
+[2025-07-05 04:37:21,441][oumi][rank0][pid:2619050][MainThread][INFO]][device_utils.py:297] GPU Metrics After Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=33, fan_speed=None, fan_speeds=None, power_usage_watts=134.097, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:21,442][oumi][rank0][pid:2619050][MainThread][INFO]][torch_utils.py:136] Peak GPU memory usage: 29.37 GB
+[2025-07-05 04:37:21,442][oumi][rank0][pid:2619050][MainThread][INFO]][train.py:525] Saving final state...
+[2025-07-05 04:37:21,451][oumi][rank0][pid:2619050][MainThread][INFO]][train.py:530] Saving final model...
+[2025-07-05 04:37:21,454][oumi][rank0][pid:2619050][MainThread][INFO]][hf_trainer.py:142] Saving FULL_STATE_DICT for final model checkpoint.
+[2025-07-05 04:43:01,650][oumi][rank0][pid:2619050][MainThread][INFO]][hf_trainer.py:149] Model has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:05,405][oumi][rank0][pid:2619050][MainThread][INFO]][hf_trainer.py:153] Processor config has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:07,241][oumi][rank0][pid:2619050][MainThread][INFO]][train.py:214] 
+
+» We're always looking for feedback. What's one thing we can improve? https://oumi.ai/feedback

logs/rank_0001.log

@@ -0,0 +1,40 @@
+[2025-07-04 10:06:56,067][oumi][rank1][pid:2619051][MainThread][INFO]][models.py:433] Setting tokenizer to use the 'right' padding side for model 'Qwen/Qwen2.5-VL-7B-Instruct'. The 'right' padding side is configured as the default value for this model type.
+[2025-07-04 10:06:57,882][oumi][rank1][pid:2619051][MainThread][INFO]][models.py:479] Using the chat template 'qwen2-vl-instruct' specified in model config for model 'Qwen/Qwen2.5-VL-7B-Instruct'. 
+[2025-07-04 10:07:03,033][oumi][rank1][pid:2619051][MainThread][WARNING]][data.py:66] Using torchdata preprocessing pipeline. This is currently in beta and may not be stable.
+[2025-07-04 10:07:03,033][oumi][rank1][pid:2619051][MainThread][INFO]][base_map_dataset.py:91] Creating map dataset (type: HuggingFaceVisionDataset)... dataset_name: 'penfever/MM-MathInstruct-to-r1-format-filtered'
+[2025-07-04 10:07:05,285][oumi][rank1][pid:2619051][MainThread][INFO]][base_map_dataset.py:487] Dataset Info:
+	Split: train
+	Version: 0.0.0
+	Dataset size: 57058499
+	Download size: 48789762
+	Size: 105848261 bytes
+	Rows: 4286
+	Columns: ['problem', 'solution', 'original_question', 'original_answer', 'image']
+[2025-07-04 10:07:07,699][oumi][rank1][pid:2619051][MainThread][INFO]][base_map_dataset.py:426] Loaded DataFrame with shape: (4286, 5). Columns:
+problem              object
+solution             object
+original_question    object
+original_answer      object
+image                object
+dtype: object
+[2025-07-04 10:07:21,069][oumi][rank1][pid:2619051][MainThread][INFO]][distributed.py:303] Initialized distributed (True): DeviceRankInfo(world_size=8, rank=1, local_world_size=8, local_rank=1)
+[2025-07-04 10:07:21,070][oumi][rank1][pid:2619051][MainThread][INFO]][train.py:406] Set Accelerate environment variables for FSDP: {'ACCELERATE_DYNAMO_BACKEND': 'NO', 'ACCELERATE_DYNAMO_MODE': 'default', 'ACCELERATE_DYNAMO_USE_FULLGRAPH': 'False', 'ACCELERATE_DYNAMO_USE_DYNAMIC': 'False', 'FSDP_CPU_RAM_EFFICIENT_LOADING': 'true', 'FSDP_USE_ORIG_PARAMS': 'true', 'ACCELERATE_USE_FSDP': 'true', 'FSDP_SHARDING_STRATEGY': 'HYBRID_SHARD', 'FSDP_OFFLOAD_PARAMS': 'false', 'ACCELERATE_MIXED_PRECISION': 'bf16', 'FSDP_BACKWARD_PREFETCH': 'BACKWARD_PRE', 'FSDP_FORWARD_PREFETCH': 'true', 'FSDP_STATE_DICT_TYPE': 'FULL_STATE_DICT', 'FSDP_AUTO_WRAP_POLICY': 'SIZE_BASED_WRAP', 'FSDP_MIN_NUM_PARAMS': '100000', 'FSDP_SYNC_MODULE_STATES': 'true', 'FSDP_ACTIVATION_CHECKPOINTING': 'true'}
+[2025-07-04 10:07:21,070][oumi][rank1][pid:2619051][MainThread][INFO]][models.py:219] Accelerate FSDP run detected! Setting device_map to None.
+[2025-07-04 10:07:21,071][oumi][rank1][pid:2619051][MainThread][INFO]][models.py:228] Building model using device_map: None (DeviceRankInfo(world_size=8, rank=1, local_world_size=8, local_rank=1))...
+[2025-07-04 10:07:21,100][oumi][rank1][pid:2619051][MainThread][INFO]][models.py:300] Using model class: <class 'transformers.models.auto.modeling_auto.AutoModelForVision2Seq'> to instantiate model.
+[2025-07-04 10:07:24,479][oumi][rank1][pid:2619051][MainThread][INFO]][torch_profiler_utils.py:164] PROF: Torch Profiler disabled!
+[2025-07-04 10:07:24,479][oumi][rank1][pid:2619051][MainThread][WARNING]][callbacks.py:72] MFU logging requires packed datasets. Skipping MFU callbacks.
+[2025-07-04 10:07:24,827][oumi][rank1][pid:2619051][MainThread][INFO]][device_utils.py:297] GPU Metrics Before Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=6166.0, temperature=30, fan_speed=None, fan_speeds=None, power_usage_watts=121.266, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 10:07:25,288][oumi][rank1][pid:2619051][MainThread][INFO]][train.py:510] Training init time: 29.332s
+[2025-07-04 10:07:25,288][oumi][rank1][pid:2619051][MainThread][INFO]][train.py:511] Starting training... (TrainerType.TRL_SFT, transformers: 4.51.3)
+[2025-07-05 04:37:21,400][oumi][rank1][pid:2619051][MainThread][INFO]][train.py:518] Training is Complete.
+[2025-07-05 04:37:21,406][oumi][rank1][pid:2619051][MainThread][INFO]][device_utils.py:297] GPU Metrics After Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=33, fan_speed=None, fan_speeds=None, power_usage_watts=134.097, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:21,407][oumi][rank1][pid:2619051][MainThread][INFO]][torch_utils.py:136] Peak GPU memory usage: 26.58 GB
+[2025-07-05 04:37:21,407][oumi][rank1][pid:2619051][MainThread][INFO]][train.py:525] Saving final state...
+[2025-07-05 04:37:21,453][oumi][rank1][pid:2619051][MainThread][INFO]][train.py:530] Saving final model...
+[2025-07-05 04:37:21,453][oumi][rank1][pid:2619051][MainThread][INFO]][hf_trainer.py:142] Saving FULL_STATE_DICT for final model checkpoint.
+[2025-07-05 04:40:16,622][oumi][rank1][pid:2619051][MainThread][INFO]][hf_trainer.py:149] Model has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:40:20,790][oumi][rank1][pid:2619051][MainThread][INFO]][hf_trainer.py:153] Processor config has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:06,891][oumi][rank1][pid:2619051][MainThread][INFO]][train.py:214] 
+
+» We're always looking for feedback. What's one thing we can improve? https://oumi.ai/feedback

logs/rank_0002.log

@@ -0,0 +1,40 @@
+[2025-07-04 10:06:56,062][oumi][rank2][pid:2619052][MainThread][INFO]][models.py:433] Setting tokenizer to use the 'right' padding side for model 'Qwen/Qwen2.5-VL-7B-Instruct'. The 'right' padding side is configured as the default value for this model type.
+[2025-07-04 10:06:57,926][oumi][rank2][pid:2619052][MainThread][INFO]][models.py:479] Using the chat template 'qwen2-vl-instruct' specified in model config for model 'Qwen/Qwen2.5-VL-7B-Instruct'. 
+[2025-07-04 10:07:02,953][oumi][rank2][pid:2619052][MainThread][WARNING]][data.py:66] Using torchdata preprocessing pipeline. This is currently in beta and may not be stable.
+[2025-07-04 10:07:02,953][oumi][rank2][pid:2619052][MainThread][INFO]][base_map_dataset.py:91] Creating map dataset (type: HuggingFaceVisionDataset)... dataset_name: 'penfever/MM-MathInstruct-to-r1-format-filtered'
+[2025-07-04 10:07:05,190][oumi][rank2][pid:2619052][MainThread][INFO]][base_map_dataset.py:487] Dataset Info:
+	Split: train
+	Version: 0.0.0
+	Dataset size: 57058499
+	Download size: 48789762
+	Size: 105848261 bytes
+	Rows: 4286
+	Columns: ['problem', 'solution', 'original_question', 'original_answer', 'image']
+[2025-07-04 10:07:07,667][oumi][rank2][pid:2619052][MainThread][INFO]][base_map_dataset.py:426] Loaded DataFrame with shape: (4286, 5). Columns:
+problem              object
+solution             object
+original_question    object
+original_answer      object
+image                object
+dtype: object
+[2025-07-04 10:07:21,065][oumi][rank2][pid:2619052][MainThread][INFO]][distributed.py:303] Initialized distributed (True): DeviceRankInfo(world_size=8, rank=2, local_world_size=8, local_rank=2)
+[2025-07-04 10:07:21,065][oumi][rank2][pid:2619052][MainThread][INFO]][train.py:406] Set Accelerate environment variables for FSDP: {'ACCELERATE_DYNAMO_BACKEND': 'NO', 'ACCELERATE_DYNAMO_MODE': 'default', 'ACCELERATE_DYNAMO_USE_FULLGRAPH': 'False', 'ACCELERATE_DYNAMO_USE_DYNAMIC': 'False', 'FSDP_CPU_RAM_EFFICIENT_LOADING': 'true', 'FSDP_USE_ORIG_PARAMS': 'true', 'ACCELERATE_USE_FSDP': 'true', 'FSDP_SHARDING_STRATEGY': 'HYBRID_SHARD', 'FSDP_OFFLOAD_PARAMS': 'false', 'ACCELERATE_MIXED_PRECISION': 'bf16', 'FSDP_BACKWARD_PREFETCH': 'BACKWARD_PRE', 'FSDP_FORWARD_PREFETCH': 'true', 'FSDP_STATE_DICT_TYPE': 'FULL_STATE_DICT', 'FSDP_AUTO_WRAP_POLICY': 'SIZE_BASED_WRAP', 'FSDP_MIN_NUM_PARAMS': '100000', 'FSDP_SYNC_MODULE_STATES': 'true', 'FSDP_ACTIVATION_CHECKPOINTING': 'true'}
+[2025-07-04 10:07:21,066][oumi][rank2][pid:2619052][MainThread][INFO]][models.py:219] Accelerate FSDP run detected! Setting device_map to None.
+[2025-07-04 10:07:21,066][oumi][rank2][pid:2619052][MainThread][INFO]][models.py:228] Building model using device_map: None (DeviceRankInfo(world_size=8, rank=2, local_world_size=8, local_rank=2))...
+[2025-07-04 10:07:21,094][oumi][rank2][pid:2619052][MainThread][INFO]][models.py:300] Using model class: <class 'transformers.models.auto.modeling_auto.AutoModelForVision2Seq'> to instantiate model.
+[2025-07-04 10:07:24,380][oumi][rank2][pid:2619052][MainThread][INFO]][torch_profiler_utils.py:164] PROF: Torch Profiler disabled!
+[2025-07-04 10:07:24,381][oumi][rank2][pid:2619052][MainThread][WARNING]][callbacks.py:72] MFU logging requires packed datasets. Skipping MFU callbacks.
+[2025-07-04 10:07:24,794][oumi][rank2][pid:2619052][MainThread][INFO]][device_utils.py:297] GPU Metrics Before Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=6166.0, temperature=30, fan_speed=None, fan_speeds=None, power_usage_watts=121.266, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 10:07:25,289][oumi][rank2][pid:2619052][MainThread][INFO]][train.py:510] Training init time: 29.331s
+[2025-07-04 10:07:25,289][oumi][rank2][pid:2619052][MainThread][INFO]][train.py:511] Starting training... (TrainerType.TRL_SFT, transformers: 4.51.3)
+[2025-07-05 04:37:21,430][oumi][rank2][pid:2619052][MainThread][INFO]][train.py:518] Training is Complete.
+[2025-07-05 04:37:21,437][oumi][rank2][pid:2619052][MainThread][INFO]][device_utils.py:297] GPU Metrics After Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=33, fan_speed=None, fan_speeds=None, power_usage_watts=127.816, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:21,438][oumi][rank2][pid:2619052][MainThread][INFO]][torch_utils.py:136] Peak GPU memory usage: 26.67 GB
+[2025-07-05 04:37:21,438][oumi][rank2][pid:2619052][MainThread][INFO]][train.py:525] Saving final state...
+[2025-07-05 04:37:21,453][oumi][rank2][pid:2619052][MainThread][INFO]][train.py:530] Saving final model...
+[2025-07-05 04:37:21,453][oumi][rank2][pid:2619052][MainThread][INFO]][hf_trainer.py:142] Saving FULL_STATE_DICT for final model checkpoint.
+[2025-07-05 04:40:16,620][oumi][rank2][pid:2619052][MainThread][INFO]][hf_trainer.py:149] Model has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:40:20,693][oumi][rank2][pid:2619052][MainThread][INFO]][hf_trainer.py:153] Processor config has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:06,319][oumi][rank2][pid:2619052][MainThread][INFO]][train.py:214] 
+
+» We're always looking for feedback. What's one thing we can improve? https://oumi.ai/feedback

logs/rank_0003.log

@@ -0,0 +1,40 @@
+[2025-07-04 10:06:56,066][oumi][rank3][pid:2619053][MainThread][INFO]][models.py:433] Setting tokenizer to use the 'right' padding side for model 'Qwen/Qwen2.5-VL-7B-Instruct'. The 'right' padding side is configured as the default value for this model type.
+[2025-07-04 10:06:57,878][oumi][rank3][pid:2619053][MainThread][INFO]][models.py:479] Using the chat template 'qwen2-vl-instruct' specified in model config for model 'Qwen/Qwen2.5-VL-7B-Instruct'. 
+[2025-07-04 10:07:02,977][oumi][rank3][pid:2619053][MainThread][WARNING]][data.py:66] Using torchdata preprocessing pipeline. This is currently in beta and may not be stable.
+[2025-07-04 10:07:02,978][oumi][rank3][pid:2619053][MainThread][INFO]][base_map_dataset.py:91] Creating map dataset (type: HuggingFaceVisionDataset)... dataset_name: 'penfever/MM-MathInstruct-to-r1-format-filtered'
+[2025-07-04 10:07:05,195][oumi][rank3][pid:2619053][MainThread][INFO]][base_map_dataset.py:487] Dataset Info:
+	Split: train
+	Version: 0.0.0
+	Dataset size: 57058499
+	Download size: 48789762
+	Size: 105848261 bytes
+	Rows: 4286
+	Columns: ['problem', 'solution', 'original_question', 'original_answer', 'image']
+[2025-07-04 10:07:07,990][oumi][rank3][pid:2619053][MainThread][INFO]][base_map_dataset.py:426] Loaded DataFrame with shape: (4286, 5). Columns:
+problem              object
+solution             object
+original_question    object
+original_answer      object
+image                object
+dtype: object
+[2025-07-04 10:07:21,052][oumi][rank3][pid:2619053][MainThread][INFO]][distributed.py:303] Initialized distributed (True): DeviceRankInfo(world_size=8, rank=3, local_world_size=8, local_rank=3)
+[2025-07-04 10:07:21,054][oumi][rank3][pid:2619053][MainThread][INFO]][train.py:406] Set Accelerate environment variables for FSDP: {'ACCELERATE_DYNAMO_BACKEND': 'NO', 'ACCELERATE_DYNAMO_MODE': 'default', 'ACCELERATE_DYNAMO_USE_FULLGRAPH': 'False', 'ACCELERATE_DYNAMO_USE_DYNAMIC': 'False', 'FSDP_CPU_RAM_EFFICIENT_LOADING': 'true', 'FSDP_USE_ORIG_PARAMS': 'true', 'ACCELERATE_USE_FSDP': 'true', 'FSDP_SHARDING_STRATEGY': 'HYBRID_SHARD', 'FSDP_OFFLOAD_PARAMS': 'false', 'ACCELERATE_MIXED_PRECISION': 'bf16', 'FSDP_BACKWARD_PREFETCH': 'BACKWARD_PRE', 'FSDP_FORWARD_PREFETCH': 'true', 'FSDP_STATE_DICT_TYPE': 'FULL_STATE_DICT', 'FSDP_AUTO_WRAP_POLICY': 'SIZE_BASED_WRAP', 'FSDP_MIN_NUM_PARAMS': '100000', 'FSDP_SYNC_MODULE_STATES': 'true', 'FSDP_ACTIVATION_CHECKPOINTING': 'true'}
+[2025-07-04 10:07:21,054][oumi][rank3][pid:2619053][MainThread][INFO]][models.py:219] Accelerate FSDP run detected! Setting device_map to None.
+[2025-07-04 10:07:21,054][oumi][rank3][pid:2619053][MainThread][INFO]][models.py:228] Building model using device_map: None (DeviceRankInfo(world_size=8, rank=3, local_world_size=8, local_rank=3))...
+[2025-07-04 10:07:21,084][oumi][rank3][pid:2619053][MainThread][INFO]][models.py:300] Using model class: <class 'transformers.models.auto.modeling_auto.AutoModelForVision2Seq'> to instantiate model.
+[2025-07-04 10:07:24,294][oumi][rank3][pid:2619053][MainThread][INFO]][torch_profiler_utils.py:164] PROF: Torch Profiler disabled!
+[2025-07-04 10:07:24,294][oumi][rank3][pid:2619053][MainThread][WARNING]][callbacks.py:72] MFU logging requires packed datasets. Skipping MFU callbacks.
+[2025-07-04 10:07:24,748][oumi][rank3][pid:2619053][MainThread][INFO]][device_utils.py:297] GPU Metrics Before Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=6166.0, temperature=30, fan_speed=None, fan_speeds=None, power_usage_watts=121.277, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 10:07:25,301][oumi][rank3][pid:2619053][MainThread][INFO]][train.py:510] Training init time: 29.344s
+[2025-07-04 10:07:25,301][oumi][rank3][pid:2619053][MainThread][INFO]][train.py:511] Starting training... (TrainerType.TRL_SFT, transformers: 4.51.3)
+[2025-07-05 04:37:21,335][oumi][rank3][pid:2619053][MainThread][INFO]][train.py:518] Training is Complete.
+[2025-07-05 04:37:21,341][oumi][rank3][pid:2619053][MainThread][INFO]][device_utils.py:297] GPU Metrics After Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=33, fan_speed=None, fan_speeds=None, power_usage_watts=134.097, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:21,341][oumi][rank3][pid:2619053][MainThread][INFO]][torch_utils.py:136] Peak GPU memory usage: 31.98 GB
+[2025-07-05 04:37:21,341][oumi][rank3][pid:2619053][MainThread][INFO]][train.py:525] Saving final state...
+[2025-07-05 04:37:21,453][oumi][rank3][pid:2619053][MainThread][INFO]][train.py:530] Saving final model...
+[2025-07-05 04:37:21,453][oumi][rank3][pid:2619053][MainThread][INFO]][hf_trainer.py:142] Saving FULL_STATE_DICT for final model checkpoint.
+[2025-07-05 04:40:16,620][oumi][rank3][pid:2619053][MainThread][INFO]][hf_trainer.py:149] Model has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:40:20,801][oumi][rank3][pid:2619053][MainThread][INFO]][hf_trainer.py:153] Processor config has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:06,312][oumi][rank3][pid:2619053][MainThread][INFO]][train.py:214] 
+
+» We're always looking for feedback. What's one thing we can improve? https://oumi.ai/feedback

logs/rank_0004.log

@@ -0,0 +1,40 @@
+[2025-07-04 10:06:56,052][oumi][rank4][pid:2619054][MainThread][INFO]][models.py:433] Setting tokenizer to use the 'right' padding side for model 'Qwen/Qwen2.5-VL-7B-Instruct'. The 'right' padding side is configured as the default value for this model type.
+[2025-07-04 10:06:57,909][oumi][rank4][pid:2619054][MainThread][INFO]][models.py:479] Using the chat template 'qwen2-vl-instruct' specified in model config for model 'Qwen/Qwen2.5-VL-7B-Instruct'. 
+[2025-07-04 10:07:03,045][oumi][rank4][pid:2619054][MainThread][WARNING]][data.py:66] Using torchdata preprocessing pipeline. This is currently in beta and may not be stable.
+[2025-07-04 10:07:03,045][oumi][rank4][pid:2619054][MainThread][INFO]][base_map_dataset.py:91] Creating map dataset (type: HuggingFaceVisionDataset)... dataset_name: 'penfever/MM-MathInstruct-to-r1-format-filtered'
+[2025-07-04 10:07:05,190][oumi][rank4][pid:2619054][MainThread][INFO]][base_map_dataset.py:487] Dataset Info:
+	Split: train
+	Version: 0.0.0
+	Dataset size: 57058499
+	Download size: 48789762
+	Size: 105848261 bytes
+	Rows: 4286
+	Columns: ['problem', 'solution', 'original_question', 'original_answer', 'image']
+[2025-07-04 10:07:07,804][oumi][rank4][pid:2619054][MainThread][INFO]][base_map_dataset.py:426] Loaded DataFrame with shape: (4286, 5). Columns:
+problem              object
+solution             object
+original_question    object
+original_answer      object
+image                object
+dtype: object
+[2025-07-04 10:07:21,061][oumi][rank4][pid:2619054][MainThread][INFO]][distributed.py:303] Initialized distributed (True): DeviceRankInfo(world_size=8, rank=4, local_world_size=8, local_rank=4)
+[2025-07-04 10:07:21,062][oumi][rank4][pid:2619054][MainThread][INFO]][train.py:406] Set Accelerate environment variables for FSDP: {'ACCELERATE_DYNAMO_BACKEND': 'NO', 'ACCELERATE_DYNAMO_MODE': 'default', 'ACCELERATE_DYNAMO_USE_FULLGRAPH': 'False', 'ACCELERATE_DYNAMO_USE_DYNAMIC': 'False', 'FSDP_CPU_RAM_EFFICIENT_LOADING': 'true', 'FSDP_USE_ORIG_PARAMS': 'true', 'ACCELERATE_USE_FSDP': 'true', 'FSDP_SHARDING_STRATEGY': 'HYBRID_SHARD', 'FSDP_OFFLOAD_PARAMS': 'false', 'ACCELERATE_MIXED_PRECISION': 'bf16', 'FSDP_BACKWARD_PREFETCH': 'BACKWARD_PRE', 'FSDP_FORWARD_PREFETCH': 'true', 'FSDP_STATE_DICT_TYPE': 'FULL_STATE_DICT', 'FSDP_AUTO_WRAP_POLICY': 'SIZE_BASED_WRAP', 'FSDP_MIN_NUM_PARAMS': '100000', 'FSDP_SYNC_MODULE_STATES': 'true', 'FSDP_ACTIVATION_CHECKPOINTING': 'true'}
+[2025-07-04 10:07:21,062][oumi][rank4][pid:2619054][MainThread][INFO]][models.py:219] Accelerate FSDP run detected! Setting device_map to None.
+[2025-07-04 10:07:21,062][oumi][rank4][pid:2619054][MainThread][INFO]][models.py:228] Building model using device_map: None (DeviceRankInfo(world_size=8, rank=4, local_world_size=8, local_rank=4))...
+[2025-07-04 10:07:21,090][oumi][rank4][pid:2619054][MainThread][INFO]][models.py:300] Using model class: <class 'transformers.models.auto.modeling_auto.AutoModelForVision2Seq'> to instantiate model.
+[2025-07-04 10:07:24,417][oumi][rank4][pid:2619054][MainThread][INFO]][torch_profiler_utils.py:164] PROF: Torch Profiler disabled!
+[2025-07-04 10:07:24,417][oumi][rank4][pid:2619054][MainThread][WARNING]][callbacks.py:72] MFU logging requires packed datasets. Skipping MFU callbacks.
+[2025-07-04 10:07:24,786][oumi][rank4][pid:2619054][MainThread][INFO]][device_utils.py:297] GPU Metrics Before Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=6166.0, temperature=30, fan_speed=None, fan_speeds=None, power_usage_watts=121.266, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 10:07:25,287][oumi][rank4][pid:2619054][MainThread][INFO]][train.py:510] Training init time: 29.328s
+[2025-07-04 10:07:25,287][oumi][rank4][pid:2619054][MainThread][INFO]][train.py:511] Starting training... (TrainerType.TRL_SFT, transformers: 4.51.3)
+[2025-07-05 04:37:21,286][oumi][rank4][pid:2619054][MainThread][INFO]][train.py:518] Training is Complete.
+[2025-07-05 04:37:21,293][oumi][rank4][pid:2619054][MainThread][INFO]][device_utils.py:297] GPU Metrics After Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=33, fan_speed=None, fan_speeds=None, power_usage_watts=135.197, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:21,293][oumi][rank4][pid:2619054][MainThread][INFO]][torch_utils.py:136] Peak GPU memory usage: 29.58 GB
+[2025-07-05 04:37:21,293][oumi][rank4][pid:2619054][MainThread][INFO]][train.py:525] Saving final state...
+[2025-07-05 04:37:21,451][oumi][rank4][pid:2619054][MainThread][INFO]][train.py:530] Saving final model...
+[2025-07-05 04:37:21,452][oumi][rank4][pid:2619054][MainThread][INFO]][hf_trainer.py:142] Saving FULL_STATE_DICT for final model checkpoint.
+[2025-07-05 04:40:16,617][oumi][rank4][pid:2619054][MainThread][INFO]][hf_trainer.py:149] Model has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:40:20,756][oumi][rank4][pid:2619054][MainThread][INFO]][hf_trainer.py:153] Processor config has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:06,046][oumi][rank4][pid:2619054][MainThread][INFO]][train.py:214] 
+
+» We're always looking for feedback. What's one thing we can improve? https://oumi.ai/feedback

logs/rank_0005.log

@@ -0,0 +1,40 @@
+[2025-07-04 10:06:56,055][oumi][rank5][pid:2619055][MainThread][INFO]][models.py:433] Setting tokenizer to use the 'right' padding side for model 'Qwen/Qwen2.5-VL-7B-Instruct'. The 'right' padding side is configured as the default value for this model type.
+[2025-07-04 10:06:57,908][oumi][rank5][pid:2619055][MainThread][INFO]][models.py:479] Using the chat template 'qwen2-vl-instruct' specified in model config for model 'Qwen/Qwen2.5-VL-7B-Instruct'. 
+[2025-07-04 10:07:03,051][oumi][rank5][pid:2619055][MainThread][WARNING]][data.py:66] Using torchdata preprocessing pipeline. This is currently in beta and may not be stable.
+[2025-07-04 10:07:03,051][oumi][rank5][pid:2619055][MainThread][INFO]][base_map_dataset.py:91] Creating map dataset (type: HuggingFaceVisionDataset)... dataset_name: 'penfever/MM-MathInstruct-to-r1-format-filtered'
+[2025-07-04 10:07:05,197][oumi][rank5][pid:2619055][MainThread][INFO]][base_map_dataset.py:487] Dataset Info:
+	Split: train
+	Version: 0.0.0
+	Dataset size: 57058499
+	Download size: 48789762
+	Size: 105848261 bytes
+	Rows: 4286
+	Columns: ['problem', 'solution', 'original_question', 'original_answer', 'image']
+[2025-07-04 10:07:07,706][oumi][rank5][pid:2619055][MainThread][INFO]][base_map_dataset.py:426] Loaded DataFrame with shape: (4286, 5). Columns:
+problem              object
+solution             object
+original_question    object
+original_answer      object
+image                object
+dtype: object
+[2025-07-04 10:07:21,046][oumi][rank5][pid:2619055][MainThread][INFO]][distributed.py:303] Initialized distributed (True): DeviceRankInfo(world_size=8, rank=5, local_world_size=8, local_rank=5)
+[2025-07-04 10:07:21,047][oumi][rank5][pid:2619055][MainThread][INFO]][train.py:406] Set Accelerate environment variables for FSDP: {'ACCELERATE_DYNAMO_BACKEND': 'NO', 'ACCELERATE_DYNAMO_MODE': 'default', 'ACCELERATE_DYNAMO_USE_FULLGRAPH': 'False', 'ACCELERATE_DYNAMO_USE_DYNAMIC': 'False', 'FSDP_CPU_RAM_EFFICIENT_LOADING': 'true', 'FSDP_USE_ORIG_PARAMS': 'true', 'ACCELERATE_USE_FSDP': 'true', 'FSDP_SHARDING_STRATEGY': 'HYBRID_SHARD', 'FSDP_OFFLOAD_PARAMS': 'false', 'ACCELERATE_MIXED_PRECISION': 'bf16', 'FSDP_BACKWARD_PREFETCH': 'BACKWARD_PRE', 'FSDP_FORWARD_PREFETCH': 'true', 'FSDP_STATE_DICT_TYPE': 'FULL_STATE_DICT', 'FSDP_AUTO_WRAP_POLICY': 'SIZE_BASED_WRAP', 'FSDP_MIN_NUM_PARAMS': '100000', 'FSDP_SYNC_MODULE_STATES': 'true', 'FSDP_ACTIVATION_CHECKPOINTING': 'true'}
+[2025-07-04 10:07:21,047][oumi][rank5][pid:2619055][MainThread][INFO]][models.py:219] Accelerate FSDP run detected! Setting device_map to None.
+[2025-07-04 10:07:21,047][oumi][rank5][pid:2619055][MainThread][INFO]][models.py:228] Building model using device_map: None (DeviceRankInfo(world_size=8, rank=5, local_world_size=8, local_rank=5))...
+[2025-07-04 10:07:21,083][oumi][rank5][pid:2619055][MainThread][INFO]][models.py:300] Using model class: <class 'transformers.models.auto.modeling_auto.AutoModelForVision2Seq'> to instantiate model.
+[2025-07-04 10:07:24,308][oumi][rank5][pid:2619055][MainThread][INFO]][torch_profiler_utils.py:164] PROF: Torch Profiler disabled!
+[2025-07-04 10:07:24,308][oumi][rank5][pid:2619055][MainThread][WARNING]][callbacks.py:72] MFU logging requires packed datasets. Skipping MFU callbacks.
+[2025-07-04 10:07:24,772][oumi][rank5][pid:2619055][MainThread][INFO]][device_utils.py:297] GPU Metrics Before Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=6166.0, temperature=30, fan_speed=None, fan_speeds=None, power_usage_watts=121.277, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 10:07:25,292][oumi][rank5][pid:2619055][MainThread][INFO]][train.py:510] Training init time: 29.335s
+[2025-07-04 10:07:25,292][oumi][rank5][pid:2619055][MainThread][INFO]][train.py:511] Starting training... (TrainerType.TRL_SFT, transformers: 4.51.3)
+[2025-07-05 04:37:21,368][oumi][rank5][pid:2619055][MainThread][INFO]][train.py:518] Training is Complete.
+[2025-07-05 04:37:21,373][oumi][rank5][pid:2619055][MainThread][INFO]][device_utils.py:297] GPU Metrics After Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=33, fan_speed=None, fan_speeds=None, power_usage_watts=134.097, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:21,373][oumi][rank5][pid:2619055][MainThread][INFO]][torch_utils.py:136] Peak GPU memory usage: 28.73 GB
+[2025-07-05 04:37:21,373][oumi][rank5][pid:2619055][MainThread][INFO]][train.py:525] Saving final state...
+[2025-07-05 04:37:21,454][oumi][rank5][pid:2619055][MainThread][INFO]][train.py:530] Saving final model...
+[2025-07-05 04:37:21,454][oumi][rank5][pid:2619055][MainThread][INFO]][hf_trainer.py:142] Saving FULL_STATE_DICT for final model checkpoint.
+[2025-07-05 04:40:16,620][oumi][rank5][pid:2619055][MainThread][INFO]][hf_trainer.py:149] Model has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:40:20,536][oumi][rank5][pid:2619055][MainThread][INFO]][hf_trainer.py:153] Processor config has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:05,901][oumi][rank5][pid:2619055][MainThread][INFO]][train.py:214] 
+
+» We're always looking for feedback. What's one thing we can improve? https://oumi.ai/feedback

logs/rank_0006.log

@@ -0,0 +1,40 @@
+[2025-07-04 10:06:56,070][oumi][rank6][pid:2619056][MainThread][INFO]][models.py:433] Setting tokenizer to use the 'right' padding side for model 'Qwen/Qwen2.5-VL-7B-Instruct'. The 'right' padding side is configured as the default value for this model type.
+[2025-07-04 10:06:57,902][oumi][rank6][pid:2619056][MainThread][INFO]][models.py:479] Using the chat template 'qwen2-vl-instruct' specified in model config for model 'Qwen/Qwen2.5-VL-7B-Instruct'. 
+[2025-07-04 10:07:03,025][oumi][rank6][pid:2619056][MainThread][WARNING]][data.py:66] Using torchdata preprocessing pipeline. This is currently in beta and may not be stable.
+[2025-07-04 10:07:03,026][oumi][rank6][pid:2619056][MainThread][INFO]][base_map_dataset.py:91] Creating map dataset (type: HuggingFaceVisionDataset)... dataset_name: 'penfever/MM-MathInstruct-to-r1-format-filtered'
+[2025-07-04 10:07:05,222][oumi][rank6][pid:2619056][MainThread][INFO]][base_map_dataset.py:487] Dataset Info:
+	Split: train
+	Version: 0.0.0
+	Dataset size: 57058499
+	Download size: 48789762
+	Size: 105848261 bytes
+	Rows: 4286
+	Columns: ['problem', 'solution', 'original_question', 'original_answer', 'image']
+[2025-07-04 10:07:07,724][oumi][rank6][pid:2619056][MainThread][INFO]][base_map_dataset.py:426] Loaded DataFrame with shape: (4286, 5). Columns:
+problem              object
+solution             object
+original_question    object
+original_answer      object
+image                object
+dtype: object
+[2025-07-04 10:07:21,068][oumi][rank6][pid:2619056][MainThread][INFO]][distributed.py:303] Initialized distributed (True): DeviceRankInfo(world_size=8, rank=6, local_world_size=8, local_rank=6)
+[2025-07-04 10:07:21,070][oumi][rank6][pid:2619056][MainThread][INFO]][train.py:406] Set Accelerate environment variables for FSDP: {'ACCELERATE_DYNAMO_BACKEND': 'NO', 'ACCELERATE_DYNAMO_MODE': 'default', 'ACCELERATE_DYNAMO_USE_FULLGRAPH': 'False', 'ACCELERATE_DYNAMO_USE_DYNAMIC': 'False', 'FSDP_CPU_RAM_EFFICIENT_LOADING': 'true', 'FSDP_USE_ORIG_PARAMS': 'true', 'ACCELERATE_USE_FSDP': 'true', 'FSDP_SHARDING_STRATEGY': 'HYBRID_SHARD', 'FSDP_OFFLOAD_PARAMS': 'false', 'ACCELERATE_MIXED_PRECISION': 'bf16', 'FSDP_BACKWARD_PREFETCH': 'BACKWARD_PRE', 'FSDP_FORWARD_PREFETCH': 'true', 'FSDP_STATE_DICT_TYPE': 'FULL_STATE_DICT', 'FSDP_AUTO_WRAP_POLICY': 'SIZE_BASED_WRAP', 'FSDP_MIN_NUM_PARAMS': '100000', 'FSDP_SYNC_MODULE_STATES': 'true', 'FSDP_ACTIVATION_CHECKPOINTING': 'true'}
+[2025-07-04 10:07:21,070][oumi][rank6][pid:2619056][MainThread][INFO]][models.py:219] Accelerate FSDP run detected! Setting device_map to None.
+[2025-07-04 10:07:21,071][oumi][rank6][pid:2619056][MainThread][INFO]][models.py:228] Building model using device_map: None (DeviceRankInfo(world_size=8, rank=6, local_world_size=8, local_rank=6))...
+[2025-07-04 10:07:21,106][oumi][rank6][pid:2619056][MainThread][INFO]][models.py:300] Using model class: <class 'transformers.models.auto.modeling_auto.AutoModelForVision2Seq'> to instantiate model.
+[2025-07-04 10:07:24,500][oumi][rank6][pid:2619056][MainThread][INFO]][torch_profiler_utils.py:164] PROF: Torch Profiler disabled!
+[2025-07-04 10:07:24,500][oumi][rank6][pid:2619056][MainThread][WARNING]][callbacks.py:72] MFU logging requires packed datasets. Skipping MFU callbacks.
+[2025-07-04 10:07:24,822][oumi][rank6][pid:2619056][MainThread][INFO]][device_utils.py:297] GPU Metrics Before Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=6166.0, temperature=30, fan_speed=None, fan_speeds=None, power_usage_watts=121.266, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 10:07:25,293][oumi][rank6][pid:2619056][MainThread][INFO]][train.py:510] Training init time: 29.334s
+[2025-07-04 10:07:25,293][oumi][rank6][pid:2619056][MainThread][INFO]][train.py:511] Starting training... (TrainerType.TRL_SFT, transformers: 4.51.3)
+[2025-07-05 04:37:21,401][oumi][rank6][pid:2619056][MainThread][INFO]][train.py:518] Training is Complete.
+[2025-07-05 04:37:21,402][oumi][rank6][pid:2619056][MainThread][INFO]][device_utils.py:297] GPU Metrics After Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=33, fan_speed=None, fan_speeds=None, power_usage_watts=134.097, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:21,406][oumi][rank6][pid:2619056][MainThread][INFO]][torch_utils.py:136] Peak GPU memory usage: 29.56 GB
+[2025-07-05 04:37:21,406][oumi][rank6][pid:2619056][MainThread][INFO]][train.py:525] Saving final state...
+[2025-07-05 04:37:21,452][oumi][rank6][pid:2619056][MainThread][INFO]][train.py:530] Saving final model...
+[2025-07-05 04:37:21,452][oumi][rank6][pid:2619056][MainThread][INFO]][hf_trainer.py:142] Saving FULL_STATE_DICT for final model checkpoint.
+[2025-07-05 04:40:16,619][oumi][rank6][pid:2619056][MainThread][INFO]][hf_trainer.py:149] Model has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:40:20,620][oumi][rank6][pid:2619056][MainThread][INFO]][hf_trainer.py:153] Processor config has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:06,598][oumi][rank6][pid:2619056][MainThread][INFO]][train.py:214] 
+
+» We're always looking for feedback. What's one thing we can improve? https://oumi.ai/feedback

logs/rank_0007.log

@@ -0,0 +1,40 @@
+[2025-07-04 10:06:56,058][oumi][rank7][pid:2619057][MainThread][INFO]][models.py:433] Setting tokenizer to use the 'right' padding side for model 'Qwen/Qwen2.5-VL-7B-Instruct'. The 'right' padding side is configured as the default value for this model type.
+[2025-07-04 10:06:57,900][oumi][rank7][pid:2619057][MainThread][INFO]][models.py:479] Using the chat template 'qwen2-vl-instruct' specified in model config for model 'Qwen/Qwen2.5-VL-7B-Instruct'. 
+[2025-07-04 10:07:03,049][oumi][rank7][pid:2619057][MainThread][WARNING]][data.py:66] Using torchdata preprocessing pipeline. This is currently in beta and may not be stable.
+[2025-07-04 10:07:03,049][oumi][rank7][pid:2619057][MainThread][INFO]][base_map_dataset.py:91] Creating map dataset (type: HuggingFaceVisionDataset)... dataset_name: 'penfever/MM-MathInstruct-to-r1-format-filtered'
+[2025-07-04 10:07:05,217][oumi][rank7][pid:2619057][MainThread][INFO]][base_map_dataset.py:487] Dataset Info:
+	Split: train
+	Version: 0.0.0
+	Dataset size: 57058499
+	Download size: 48789762
+	Size: 105848261 bytes
+	Rows: 4286
+	Columns: ['problem', 'solution', 'original_question', 'original_answer', 'image']
+[2025-07-04 10:07:07,979][oumi][rank7][pid:2619057][MainThread][INFO]][base_map_dataset.py:426] Loaded DataFrame with shape: (4286, 5). Columns:
+problem              object
+solution             object
+original_question    object
+original_answer      object
+image                object
+dtype: object
+[2025-07-04 10:07:21,057][oumi][rank7][pid:2619057][MainThread][INFO]][distributed.py:303] Initialized distributed (True): DeviceRankInfo(world_size=8, rank=7, local_world_size=8, local_rank=7)
+[2025-07-04 10:07:21,058][oumi][rank7][pid:2619057][MainThread][INFO]][train.py:406] Set Accelerate environment variables for FSDP: {'ACCELERATE_DYNAMO_BACKEND': 'NO', 'ACCELERATE_DYNAMO_MODE': 'default', 'ACCELERATE_DYNAMO_USE_FULLGRAPH': 'False', 'ACCELERATE_DYNAMO_USE_DYNAMIC': 'False', 'FSDP_CPU_RAM_EFFICIENT_LOADING': 'true', 'FSDP_USE_ORIG_PARAMS': 'true', 'ACCELERATE_USE_FSDP': 'true', 'FSDP_SHARDING_STRATEGY': 'HYBRID_SHARD', 'FSDP_OFFLOAD_PARAMS': 'false', 'ACCELERATE_MIXED_PRECISION': 'bf16', 'FSDP_BACKWARD_PREFETCH': 'BACKWARD_PRE', 'FSDP_FORWARD_PREFETCH': 'true', 'FSDP_STATE_DICT_TYPE': 'FULL_STATE_DICT', 'FSDP_AUTO_WRAP_POLICY': 'SIZE_BASED_WRAP', 'FSDP_MIN_NUM_PARAMS': '100000', 'FSDP_SYNC_MODULE_STATES': 'true', 'FSDP_ACTIVATION_CHECKPOINTING': 'true'}
+[2025-07-04 10:07:21,058][oumi][rank7][pid:2619057][MainThread][INFO]][models.py:219] Accelerate FSDP run detected! Setting device_map to None.
+[2025-07-04 10:07:21,058][oumi][rank7][pid:2619057][MainThread][INFO]][models.py:228] Building model using device_map: None (DeviceRankInfo(world_size=8, rank=7, local_world_size=8, local_rank=7))...
+[2025-07-04 10:07:21,086][oumi][rank7][pid:2619057][MainThread][INFO]][models.py:300] Using model class: <class 'transformers.models.auto.modeling_auto.AutoModelForVision2Seq'> to instantiate model.
+[2025-07-04 10:07:24,262][oumi][rank7][pid:2619057][MainThread][INFO]][torch_profiler_utils.py:164] PROF: Torch Profiler disabled!
+[2025-07-04 10:07:24,262][oumi][rank7][pid:2619057][MainThread][WARNING]][callbacks.py:72] MFU logging requires packed datasets. Skipping MFU callbacks.
+[2025-07-04 10:07:24,740][oumi][rank7][pid:2619057][MainThread][INFO]][device_utils.py:297] GPU Metrics Before Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=6166.0, temperature=30, fan_speed=None, fan_speeds=None, power_usage_watts=121.277, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-04 10:07:25,299][oumi][rank7][pid:2619057][MainThread][INFO]][train.py:510] Training init time: 29.341s
+[2025-07-04 10:07:25,299][oumi][rank7][pid:2619057][MainThread][INFO]][train.py:511] Starting training... (TrainerType.TRL_SFT, transformers: 4.51.3)
+[2025-07-05 04:37:21,399][oumi][rank7][pid:2619057][MainThread][INFO]][train.py:518] Training is Complete.
+[2025-07-05 04:37:21,401][oumi][rank7][pid:2619057][MainThread][INFO]][device_utils.py:297] GPU Metrics After Training: GPU runtime info: NVidiaGpuRuntimeInfo(device_index=0, device_count=8, used_memory_mb=34792.0, temperature=33, fan_speed=None, fan_speeds=None, power_usage_watts=134.097, power_limit_watts=700.0, gpu_utilization=0, memory_utilization=0, performance_state=0, clock_speed_graphics=1980, clock_speed_sm=1980, clock_speed_memory=2619).
+[2025-07-05 04:37:21,402][oumi][rank7][pid:2619057][MainThread][INFO]][torch_utils.py:136] Peak GPU memory usage: 30.10 GB
+[2025-07-05 04:37:21,402][oumi][rank7][pid:2619057][MainThread][INFO]][train.py:525] Saving final state...
+[2025-07-05 04:37:21,454][oumi][rank7][pid:2619057][MainThread][INFO]][train.py:530] Saving final model...
+[2025-07-05 04:37:21,454][oumi][rank7][pid:2619057][MainThread][INFO]][hf_trainer.py:142] Saving FULL_STATE_DICT for final model checkpoint.
+[2025-07-05 04:40:16,621][oumi][rank7][pid:2619057][MainThread][INFO]][hf_trainer.py:149] Model has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:40:20,698][oumi][rank7][pid:2619057][MainThread][INFO]][hf_trainer.py:153] Processor config has been saved at output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+[2025-07-05 04:43:06,529][oumi][rank7][pid:2619057][MainThread][INFO]][train.py:214] 
+
+» We're always looking for feedback. What's one thing we can improve? https://oumi.ai/feedback

model-00001-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25fbedaea8c11fdb479dba6c19b80b01f5d01b4986237aa3c1409183039395eb
+size 4952311608

model-00002-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f56ab8b5118155e68ee8b476ae7da83c842862316009d899f14c1fce1fa2b2a
+size 4984124272

model-00003-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d206ad218950e56775aef51a9c6f671364fef793e0b2458ea725fa11f00a4aed
+size 4932743936

model-00004-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1799f2b014c2734d7a51285c129bf4b8917f2d19a3ac5d9f202be5fde73ecb5
+size 4998852296

model-00005-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67b37cd470db735b247d9bcfba76d012667a622750a5887fdc886bff5c148e53
+size 4984124336

model-00006-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f58c58bf796f45e1d33f0ee102b1f72031df2727d9c02f0eac860feca1b13a7
+size 4932743992

model-00007-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcaec0e85ec6b051d57e263aff6bd7f27926b79689a40086a5da56deb573b37d
+size 3383846800

model.safetensors.index.json

@@ -0,0 +1,736 @@
+{
+  "metadata": {
+    "total_size": 33168666624
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00007-of-00007.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00007-of-00007.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00007-of-00007.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00007-of-00007.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00007-of-00007.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.norm.weight": "model-00007-of-00007.safetensors",
+    "visual.blocks.0.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.0.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.1.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.10.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.11.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.12.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.13.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.14.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.15.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.16.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.17.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.18.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.19.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.2.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.20.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.21.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.22.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.23.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.24.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.25.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.26.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.27.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.28.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.29.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.3.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.30.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.31.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.4.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.5.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.6.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.7.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.8.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.attn.proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.attn.proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.attn.qkv.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.attn.qkv.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.mlp.down_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.mlp.gate_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.mlp.up_proj.bias": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.norm1.weight": "model-00001-of-00007.safetensors",
+    "visual.blocks.9.norm2.weight": "model-00001-of-00007.safetensors",
+    "visual.merger.ln_q.weight": "model-00001-of-00007.safetensors",
+    "visual.merger.mlp.0.bias": "model-00001-of-00007.safetensors",
+    "visual.merger.mlp.0.weight": "model-00001-of-00007.safetensors",
+    "visual.merger.mlp.2.bias": "model-00001-of-00007.safetensors",
+    "visual.merger.mlp.2.weight": "model-00001-of-00007.safetensors",
+    "visual.patch_embed.proj.weight": "model-00001-of-00007.safetensors"
+  }
+}

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 12845056,
+    "shortest_edge": 3136
+  },
+  "temporal_patch_size": 2
+}

runs/Jul04_10-07-24_oumi-compute004/events.out.tfevents.1751623712.oumi-compute004.2619050.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a7a1178cb9df90d8edd3eea89965c5626528e98d603bc8da169ea35c9b65695
+size 36575

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

telemetry/devices_info.txt

@@ -0,0 +1,9 @@
+CPU cores: 208 CUDA devices: 8
+device(0)='NVIDIA H100 80GB HBM3' Capability: (9, 0) Memory: [Total: 79.19GiB Free: 75.08GiB Allocated: 0.0GiB Cached: 0.0GiB]
+device(1)='NVIDIA H100 80GB HBM3' Capability: (9, 0) Memory: [Total: 79.19GiB Free: 78.67GiB Allocated: 0.0GiB Cached: 0.0GiB]
+device(2)='NVIDIA H100 80GB HBM3' Capability: (9, 0) Memory: [Total: 79.19GiB Free: 78.67GiB Allocated: 0.0GiB Cached: 0.0GiB]
+device(3)='NVIDIA H100 80GB HBM3' Capability: (9, 0) Memory: [Total: 79.19GiB Free: 78.67GiB Allocated: 0.0GiB Cached: 0.0GiB]
+device(4)='NVIDIA H100 80GB HBM3' Capability: (9, 0) Memory: [Total: 79.19GiB Free: 78.67GiB Allocated: 0.0GiB Cached: 0.0GiB]
+device(5)='NVIDIA H100 80GB HBM3' Capability: (9, 0) Memory: [Total: 79.19GiB Free: 78.67GiB Allocated: 0.0GiB Cached: 0.0GiB]
+device(6)='NVIDIA H100 80GB HBM3' Capability: (9, 0) Memory: [Total: 79.19GiB Free: 78.67GiB Allocated: 0.0GiB Cached: 0.0GiB]
+device(7)='NVIDIA H100 80GB HBM3' Capability: (9, 0) Memory: [Total: 79.19GiB Free: 78.67GiB Allocated: 0.0GiB Cached: 0.0GiB]

telemetry/telemetry_callback_metrics_rank0000.json

@@ -0,0 +1,10 @@
+{
+  "train_runtime": 66532.5206,
+  "train_samples_per_second": 0.451,
+  "train_steps_per_second": 0.056,
+  "train_tokens_per_second": 70.398,
+  "total_flos": 1.4113858770069094e+17,
+  "train_loss": 0.21003443336486816,
+  "epoch": 6.996268656716418,
+  "num_input_tokens_seen": 24290780
+}

telemetry/telemetry_callback_rank0000.json

@@ -0,0 +1,39 @@
+{
+  "hostname": "oumi-compute004",
+  "total_time": 66595.79496112792,
+  "timers": {
+    "epochs": {
+      "count": 7.0,
+      "mean": 9503.914948322012,
+      "median": 9508.863333210349,
+      "std_dev": 24.585813027685713,
+      "min": 9471.0654975418,
+      "max": 9534.607692892198,
+      "total": 66527.40463825408,
+      "percentage": 99.89730534350741
+    },
+    "microsteps": {
+      "count": 3748.0,
+      "mean": 17.662540777144716,
+      "median": 17.44136324687861,
+      "std_dev": 1.5278314346422193,
+      "min": 15.953362683299929,
+      "max": 25.46622396213934,
+      "total": 66199.2028327384,
+      "percentage": 99.40447872328726
+    },
+    "steps": {
+      "count": 3748.0,
+      "mean": 17.662547419246746,
+      "median": 17.441364587983117,
+      "std_dev": 1.5278260400852177,
+      "min": 15.953363878186792,
+      "max": 25.46622501220554,
+      "total": 66199.22772733681,
+      "percentage": 99.40451610492437
+    }
+  },
+  "cuda_timers": {},
+  "gpu_memory": [],
+  "gpu_temperature": {}
+}

telemetry/telemetry_callback_wandb_rank0000.json

@@ -0,0 +1,5 @@
+{
+  "id": "p2r4ua3y",
+  "name": "output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered",
+  "url": "https://wandb.ai/nyu-dice-lab/huggingface/runs/p2r4ua3y"
+}

telemetry/training_config.yaml

@@ -0,0 +1,206 @@
+data:
+  train:
+    datasets:
+    - dataset_name: hf_vision
+      dataset_path: null
+      subset: null
+      split: train
+      dataset_kwargs:
+        hf_dataset_path: penfever/MM-MathInstruct-to-r1-format-filtered
+        image_column: image
+        question_column: problem
+        answer_column: solution
+        return_tensors: true
+        processor_name: Qwen/Qwen2.5-VL-7B-Instruct
+        return_conversations: true
+      sample_count: null
+      mixture_proportion: null
+      shuffle: true
+      seed: 42
+      shuffle_buffer_size: 1000
+      trust_remote_code: true
+      transform_num_workers: auto
+    collator_name: vision_language_sft
+    collator_kwargs:
+      process_individually: true
+    pack: false
+    stream: false
+    target_col: null
+    mixture_strategy: first_exhausted
+    seed: null
+    use_async_dataset: false
+    use_torchdata: true
+  test:
+    datasets: []
+    collator_name: null
+    collator_kwargs: {}
+    pack: false
+    stream: false
+    target_col: null
+    mixture_strategy: first_exhausted
+    seed: null
+    use_async_dataset: false
+    use_torchdata: null
+  validation:
+    datasets: []
+    collator_name: null
+    collator_kwargs: {}
+    pack: false
+    stream: false
+    target_col: null
+    mixture_strategy: first_exhausted
+    seed: null
+    use_async_dataset: false
+    use_torchdata: null
+model:
+  model_name: Qwen/Qwen2.5-VL-7B-Instruct
+  adapter_model: null
+  tokenizer_name: null
+  tokenizer_pad_token: null
+  tokenizer_kwargs: {}
+  processor_kwargs: {}
+  model_max_length: 10000
+  load_pretrained_weights: true
+  trust_remote_code: true
+  torch_dtype_str: bfloat16
+  compile: false
+  chat_template: qwen2-vl-instruct
+  attn_implementation: sdpa
+  device_map: auto
+  model_kwargs: {}
+  enable_liger_kernel: false
+  shard_for_eval: false
+  freeze_layers: []
+  model_revision: null
+training:
+  use_peft: false
+  trainer_type: TRL_SFT
+  enable_gradient_checkpointing: true
+  gradient_checkpointing_kwargs:
+    use_reentrant: false
+  output_dir: output/qwen2_5_vl_7b_MM-MathInstruct-to-r1-format-filtered
+  per_device_train_batch_size: 1
+  per_device_eval_batch_size: 8
+  gradient_accumulation_steps: 1
+  max_steps: 3750
+  num_train_epochs: 5
+  save_epoch: false
+  save_steps: 0
+  save_final_model: true
+  seed: 42
+  data_seed: 42
+  use_deterministic: false
+  full_determinism: false
+  run_name: null
+  metrics_function: null
+  reward_functions: null
+  grpo:
+    model_init_kwargs: {}
+    max_prompt_length: null
+    max_completion_length: null
+    num_generations: null
+    temperature: 0.9
+    remove_unused_columns: false
+    repetition_penalty: 1.0
+    use_vllm: false
+    vllm_device: null
+    vllm_gpu_memory_utilization: 0.9
+    vllm_dtype: null
+    vllm_max_model_len: null
+    epsilon: 0.2
+    log_completions: false
+  log_level: info
+  dep_log_level: warning
+  enable_wandb: true
+  enable_mlflow: false
+  enable_tensorboard: true
+  logging_strategy: steps
+  logging_dir: null
+  logging_steps: 50
+  logging_first_step: false
+  eval_strategy: 'no'
+  eval_steps: 500
+  learning_rate: 2.0e-05
+  lr_scheduler_type: cosine
+  lr_scheduler_kwargs: {}
+  warmup_ratio: 0.03
+  warmup_steps: null
+  optimizer: adamw_torch_fused
+  weight_decay: 0.01
+  adam_beta1: 0.9
+  adam_beta2: 0.999
+  adam_epsilon: 1.0e-08
+  sgd_momentum: 0.0
+  mixed_precision_dtype: NONE
+  compile: false
+  include_performance_metrics: true
+  include_alternative_mfu_metrics: false
+  log_model_summary: false
+  resume_from_checkpoint: null
+  try_resume_from_last_checkpoint: false
+  dataloader_num_workers: 2
+  dataloader_persistent_workers: false
+  dataloader_prefetch_factor: 8
+  dataloader_main_process_only: false
+  ddp_find_unused_parameters: false
+  max_grad_norm: 1.0
+  trainer_kwargs:
+    max_seq_length: 10000
+    remove_unused_columns: false
+    dataset_kwargs:
+      skip_prepare_dataset: true
+  verl_config_overrides: {}
+  profiler:
+    save_dir: null
+    enable_cpu_profiling: false
+    enable_cuda_profiling: false
+    record_shapes: false
+    profile_memory: false
+    with_stack: false
+    with_flops: false
+    with_modules: false
+    row_limit: 50
+    schedule:
+      enable_schedule: false
+      wait: 0
+      warmup: 1
+      active: 3
+      repeat: 1
+      skip_first: 1
+  telemetry:
+    telemetry_dir: telemetry
+    collect_telemetry_for_all_ranks: false
+    track_gpu_temperature: false
+  empty_device_cache_steps: 1
+  nccl_default_timeout_minutes: null
+  label_ignore_index: null
+peft:
+  lora_r: 8
+  lora_alpha: 8
+  lora_dropout: 0.0
+  lora_target_modules: null
+  lora_modules_to_save: null
+  lora_bias: none
+  lora_init_weights: DEFAULT
+  lora_task_type: CAUSAL_LM
+  q_lora: false
+  q_lora_bits: 4
+  bnb_4bit_quant_type: fp4
+  llm_int8_skip_modules: null
+  use_bnb_nested_quant: false
+  bnb_4bit_quant_storage: uint8
+  bnb_4bit_compute_dtype: float32
+  peft_save_mode: ADAPTER_ONLY
+fsdp:
+  enable_fsdp: true
+  sharding_strategy: HYBRID_SHARD
+  cpu_offload: false
+  mixed_precision: bf16
+  backward_prefetch: BACKWARD_PRE
+  forward_prefetch: true
+  use_orig_params: null
+  state_dict_type: FULL_STATE_DICT
+  auto_wrap_policy: SIZE_BASED_WRAP
+  min_num_params: 100000
+  transformer_layer_cls: null
+  sync_module_states: true

telemetry/world_size.json

@@ -0,0 +1,4 @@
+{
+  "LOCAL_WORLD_SIZE": 8,
+  "WORLD_SIZE": 8
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -0,0 +1,210 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{%- for message in messages -%}{%- if loop.first and message['role'] != 'system' -%}{{ '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}{%- endif -%}{{ '<|im_start|>' + message['role'] + '\\n' }}{%- if message['content'] is string -%}{{- message['content'] -}}{%- elif message['content'] is iterable -%}{%- for item in message['content'] -%}{%- if item['type'].startswith('image') -%}{%- set image_count.value = image_count.value + 1 -%}{%- if add_vision_id -%}{{ 'Picture ' + image_count.value + ': ' }}{%- endif -%}{{ '<|vision_start|><|image_pad|><|vision_end|>' }}{%- elif item['type'].startswith('video') -%}{%- set video_count.value = video_count.value + 1 -%}{%- if add_vision_id -%}{{ 'Video ' + video_count.value + ': ' }}{%- endif -%}{{ '<|vision_start|><|video_pad|><|vision_end|>' }}{%- elif item['type']=='text' -%}{{- item['text'] if 'text' in item else item['content'] -}}{%- endif -%}{%- endfor -%}{%- endif -%}{{ '<|im_end|>\\n' }}{%- endfor -%}{%- if add_generation_prompt -%}{{- '<|im_start|>assistant\\n' -}}{%- endif -%}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 10000,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2_5_VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,795 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.996268656716418,
+  "eval_steps": 500,
+  "global_step": 3750,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09328358208955224,
+      "grad_norm": 2.4583044052124023,
+      "learning_rate": 8.672566371681418e-06,
+      "loss": 0.8247,
+      "mean_token_accuracy": 0.8019651556015015,
+      "num_input_tokens_seen": 323038,
+      "num_tokens": 323038.0,
+      "step": 50
+    },
+    {
+      "epoch": 0.1865671641791045,
+      "grad_norm": 2.1223530769348145,
+      "learning_rate": 1.7522123893805313e-05,
+      "loss": 0.6247,
+      "mean_token_accuracy": 0.8224787962436676,
+      "num_input_tokens_seen": 642936,
+      "num_tokens": 642936.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.2798507462686567,
+      "grad_norm": 5.335413932800293,
+      "learning_rate": 1.9995165482321775e-05,
+      "loss": 0.6278,
+      "mean_token_accuracy": 0.8260715854167938,
+      "num_input_tokens_seen": 976988,
+      "num_tokens": 976988.0,
+      "step": 150
+    },
+    {
+      "epoch": 0.373134328358209,
+      "grad_norm": 5.00244140625,
+      "learning_rate": 1.9972420885061576e-05,
+      "loss": 0.6247,
+      "mean_token_accuracy": 0.8278178679943085,
+      "num_input_tokens_seen": 1303490,
+      "num_tokens": 1303490.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.4664179104477612,
+      "grad_norm": 5.494905948638916,
+      "learning_rate": 1.9931077431357095e-05,
+      "loss": 0.649,
+      "mean_token_accuracy": 0.8199629688262939,
+      "num_input_tokens_seen": 1633598,
+      "num_tokens": 1633598.0,
+      "step": 250
+    },
+    {
+      "epoch": 0.5597014925373134,
+      "grad_norm": 5.7639923095703125,
+      "learning_rate": 1.9871212227957962e-05,
+      "loss": 0.6276,
+      "mean_token_accuracy": 0.8245489084720612,
+      "num_input_tokens_seen": 1959940,
+      "num_tokens": 1959940.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.6529850746268657,
+      "grad_norm": 5.357442378997803,
+      "learning_rate": 1.979293692521837e-05,
+      "loss": 0.6308,
+      "mean_token_accuracy": 0.8240770590305329,
+      "num_input_tokens_seen": 2273744,
+      "num_tokens": 2273744.0,
+      "step": 350
+    },
+    {
+      "epoch": 0.746268656716418,
+      "grad_norm": 7.147578239440918,
+      "learning_rate": 1.9696397508865917e-05,
+      "loss": 0.6753,
+      "mean_token_accuracy": 0.8150083267688751,
+      "num_input_tokens_seen": 2594534,
+      "num_tokens": 2594534.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.8395522388059702,
+      "grad_norm": 5.7040605545043945,
+      "learning_rate": 1.9581774027733947e-05,
+      "loss": 0.6221,
+      "mean_token_accuracy": 0.8259347748756408,
+      "num_input_tokens_seen": 2913982,
+      "num_tokens": 2913982.0,
+      "step": 450
+    },
+    {
+      "epoch": 0.9328358208955224,
+      "grad_norm": 5.201114654541016,
+      "learning_rate": 1.944928025796521e-05,
+      "loss": 0.6178,
+      "mean_token_accuracy": 0.8253031682968139,
+      "num_input_tokens_seen": 3233948,
+      "num_tokens": 3233948.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.0261194029850746,
+      "grad_norm": 5.712339878082275,
+      "learning_rate": 1.929916330431312e-05,
+      "loss": 0.6321,
+      "mean_token_accuracy": 0.8240388989448547,
+      "num_input_tokens_seen": 3568746,
+      "num_tokens": 3568746.0,
+      "step": 550
+    },
+    {
+      "epoch": 1.1194029850746268,
+      "grad_norm": 3.397759199142456,
+      "learning_rate": 1.9131703139284143e-05,
+      "loss": 0.4505,
+      "mean_token_accuracy": 0.8675171172618866,
+      "num_input_tokens_seen": 3886940,
+      "num_tokens": 3886940.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.212686567164179,
+      "grad_norm": 3.1189541816711426,
+      "learning_rate": 1.894721208098092e-05,
+      "loss": 0.3317,
+      "mean_token_accuracy": 0.9006148743629455,
+      "num_input_tokens_seen": 4206696,
+      "num_tokens": 4206696.0,
+      "step": 650
+    },
+    {
+      "epoch": 1.3059701492537314,
+      "grad_norm": 2.917012929916382,
+      "learning_rate": 1.874603421061986e-05,
+      "loss": 0.3288,
+      "mean_token_accuracy": 0.900334278345108,
+      "num_input_tokens_seen": 4538022,
+      "num_tokens": 4538022.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.3992537313432836,
+      "grad_norm": 3.822690963745117,
+      "learning_rate": 1.852854473080961e-05,
+      "loss": 0.4135,
+      "mean_token_accuracy": 0.8779956555366516,
+      "num_input_tokens_seen": 4868192,
+      "num_tokens": 4868192.0,
+      "step": 750
+    },
+    {
+      "epoch": 1.4925373134328357,
+      "grad_norm": 3.944916009902954,
+      "learning_rate": 1.8295149265787224e-05,
+      "loss": 0.4413,
+      "mean_token_accuracy": 0.8670336186885834,
+      "num_input_tokens_seen": 5201704,
+      "num_tokens": 5201704.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.585820895522388,
+      "grad_norm": 4.3602423667907715,
+      "learning_rate": 1.8046283104917116e-05,
+      "loss": 0.4167,
+      "mean_token_accuracy": 0.87647913813591,
+      "num_input_tokens_seen": 5521682,
+      "num_tokens": 5521682.0,
+      "step": 850
+    },
+    {
+      "epoch": 1.6791044776119404,
+      "grad_norm": 4.468040943145752,
+      "learning_rate": 1.7782410390863664e-05,
+      "loss": 0.4282,
+      "mean_token_accuracy": 0.8733036065101624,
+      "num_input_tokens_seen": 5840636,
+      "num_tokens": 5840636.0,
+      "step": 900
+    },
+    {
+      "epoch": 1.7723880597014925,
+      "grad_norm": 5.0916666984558105,
+      "learning_rate": 1.750402325395156e-05,
+      "loss": 0.422,
+      "mean_token_accuracy": 0.8751583611965179,
+      "num_input_tokens_seen": 6152288,
+      "num_tokens": 6152288.0,
+      "step": 950
+    },
+    {
+      "epoch": 1.8656716417910446,
+      "grad_norm": 3.583630084991455,
+      "learning_rate": 1.7211640894328413e-05,
+      "loss": 0.4014,
+      "mean_token_accuracy": 0.8789325177669525,
+      "num_input_tokens_seen": 6475382,
+      "num_tokens": 6475382.0,
+      "step": 1000
+    },
+    {
+      "epoch": 1.9589552238805972,
+      "grad_norm": 4.270807266235352,
+      "learning_rate": 1.6905808613641233e-05,
+      "loss": 0.4074,
+      "mean_token_accuracy": 0.8787497889995575,
+      "num_input_tokens_seen": 6799610,
+      "num_tokens": 6799610.0,
+      "step": 1050
+    },
+    {
+      "epoch": 2.0522388059701493,
+      "grad_norm": 3.991750717163086,
+      "learning_rate": 1.6587096798032984e-05,
+      "loss": 0.3727,
+      "mean_token_accuracy": 0.8879767000675202,
+      "num_input_tokens_seen": 7129112,
+      "num_tokens": 7129112.0,
+      "step": 1100
+    },
+    {
+      "epoch": 2.1455223880597014,
+      "grad_norm": 2.423408269882202,
+      "learning_rate": 1.625609985435571e-05,
+      "loss": 0.2241,
+      "mean_token_accuracy": 0.9320108902454376,
+      "num_input_tokens_seen": 7449332,
+      "num_tokens": 7449332.0,
+      "step": 1150
+    },
+    {
+      "epoch": 2.2388059701492535,
+      "grad_norm": 1.7369468212127686,
+      "learning_rate": 1.59134351015844e-05,
+      "loss": 0.1621,
+      "mean_token_accuracy": 0.9507821369171142,
+      "num_input_tokens_seen": 7773514,
+      "num_tokens": 7773514.0,
+      "step": 1200
+    },
+    {
+      "epoch": 2.332089552238806,
+      "grad_norm": 1.9339622259140015,
+      "learning_rate": 1.555974161949906e-05,
+      "loss": 0.174,
+      "mean_token_accuracy": 0.9477302300930023,
+      "num_input_tokens_seen": 8102664,
+      "num_tokens": 8102664.0,
+      "step": 1250
+    },
+    {
+      "epoch": 2.425373134328358,
+      "grad_norm": 2.86017107963562,
+      "learning_rate": 1.519567905678223e-05,
+      "loss": 0.2275,
+      "mean_token_accuracy": 0.9304351592063904,
+      "num_input_tokens_seen": 8432668,
+      "num_tokens": 8432668.0,
+      "step": 1300
+    },
+    {
+      "epoch": 2.5186567164179103,
+      "grad_norm": 2.300647735595703,
+      "learning_rate": 1.4821926400754915e-05,
+      "loss": 0.2328,
+      "mean_token_accuracy": 0.9280073237419129,
+      "num_input_tokens_seen": 8759670,
+      "num_tokens": 8759670.0,
+      "step": 1350
+    },
+    {
+      "epoch": 2.611940298507463,
+      "grad_norm": 2.8485522270202637,
+      "learning_rate": 1.4439180711045395e-05,
+      "loss": 0.2274,
+      "mean_token_accuracy": 0.9305808675289154,
+      "num_input_tokens_seen": 9083690,
+      "num_tokens": 9083690.0,
+      "step": 1400
+    },
+    {
+      "epoch": 2.705223880597015,
+      "grad_norm": 2.8772568702697754,
+      "learning_rate": 1.4048155819552617e-05,
+      "loss": 0.2385,
+      "mean_token_accuracy": 0.9277240431308746,
+      "num_input_tokens_seen": 9400762,
+      "num_tokens": 9400762.0,
+      "step": 1450
+    },
+    {
+      "epoch": 2.798507462686567,
+      "grad_norm": 3.3837602138519287,
+      "learning_rate": 1.3649580999128871e-05,
+      "loss": 0.2225,
+      "mean_token_accuracy": 0.9319508814811707,
+      "num_input_tokens_seen": 9719416,
+      "num_tokens": 9719416.0,
+      "step": 1500
+    },
+    {
+      "epoch": 2.8917910447761193,
+      "grad_norm": 3.256094217300415,
+      "learning_rate": 1.3244199603464581e-05,
+      "loss": 0.2307,
+      "mean_token_accuracy": 0.9294045794010163,
+      "num_input_tokens_seen": 10038208,
+      "num_tokens": 10038208.0,
+      "step": 1550
+    },
+    {
+      "epoch": 2.9850746268656714,
+      "grad_norm": 2.5858240127563477,
+      "learning_rate": 1.2832767680711941e-05,
+      "loss": 0.2196,
+      "mean_token_accuracy": 0.9315053272247314,
+      "num_input_tokens_seen": 10366444,
+      "num_tokens": 10366444.0,
+      "step": 1600
+    },
+    {
+      "epoch": 3.078358208955224,
+      "grad_norm": 1.8756778240203857,
+      "learning_rate": 1.2416052563433043e-05,
+      "loss": 0.19,
+      "mean_token_accuracy": 0.9421073424816132,
+      "num_input_tokens_seen": 10687314,
+      "num_tokens": 10687314.0,
+      "step": 1650
+    },
+    {
+      "epoch": 3.171641791044776,
+      "grad_norm": 0.8857208490371704,
+      "learning_rate": 1.1994831437502172e-05,
+      "loss": 0.1102,
+      "mean_token_accuracy": 0.9673730087280273,
+      "num_input_tokens_seen": 11009694,
+      "num_tokens": 11009694.0,
+      "step": 1700
+    },
+    {
+      "epoch": 3.264925373134328,
+      "grad_norm": 1.2341127395629883,
+      "learning_rate": 1.1569889892631488e-05,
+      "loss": 0.0797,
+      "mean_token_accuracy": 0.9754443645477295,
+      "num_input_tokens_seen": 11337744,
+      "num_tokens": 11337744.0,
+      "step": 1750
+    },
+    {
+      "epoch": 3.3582089552238807,
+      "grad_norm": 1.431805968284607,
+      "learning_rate": 1.1142020457223195e-05,
+      "loss": 0.0912,
+      "mean_token_accuracy": 0.9724087870121002,
+      "num_input_tokens_seen": 11665956,
+      "num_tokens": 11665956.0,
+      "step": 1800
+    },
+    {
+      "epoch": 3.451492537313433,
+      "grad_norm": 1.9885361194610596,
+      "learning_rate": 1.0712021120280951e-05,
+      "loss": 0.1156,
+      "mean_token_accuracy": 0.9630080580711364,
+      "num_input_tokens_seen": 12000080,
+      "num_tokens": 12000080.0,
+      "step": 1850
+    },
+    {
+      "epoch": 3.544776119402985,
+      "grad_norm": 1.1434038877487183,
+      "learning_rate": 1.028069384313702e-05,
+      "loss": 0.1109,
+      "mean_token_accuracy": 0.9657398784160613,
+      "num_input_tokens_seen": 12322044,
+      "num_tokens": 12322044.0,
+      "step": 1900
+    },
+    {
+      "epoch": 3.638059701492537,
+      "grad_norm": 1.1466773748397827,
+      "learning_rate": 9.848843063770963e-06,
+      "loss": 0.1098,
+      "mean_token_accuracy": 0.966309015750885,
+      "num_input_tokens_seen": 12642090,
+      "num_tokens": 12642090.0,
+      "step": 1950
+    },
+    {
+      "epoch": 3.7313432835820897,
+      "grad_norm": 1.5840574502944946,
+      "learning_rate": 9.41727419650929e-06,
+      "loss": 0.121,
+      "mean_token_accuracy": 0.961807359457016,
+      "num_input_tokens_seen": 12961610,
+      "num_tokens": 12961610.0,
+      "step": 2000
+    },
+    {
+      "epoch": 3.824626865671642,
+      "grad_norm": 1.5495262145996094,
+      "learning_rate": 8.986792129904186e-06,
+      "loss": 0.1102,
+      "mean_token_accuracy": 0.9657205975055695,
+      "num_input_tokens_seen": 13279862,
+      "num_tokens": 13279862.0,
+      "step": 2050
+    },
+    {
+      "epoch": 3.917910447761194,
+      "grad_norm": 1.7342835664749146,
+      "learning_rate": 8.558199725592856e-06,
+      "loss": 0.1156,
+      "mean_token_accuracy": 0.9632673525810241,
+      "num_input_tokens_seen": 13597816,
+      "num_tokens": 13597816.0,
+      "step": 2100
+    },
+    {
+      "epoch": 4.0111940298507465,
+      "grad_norm": 2.0673277378082275,
+      "learning_rate": 8.132296320937085e-06,
+      "loss": 0.118,
+      "mean_token_accuracy": 0.9625415456295013,
+      "num_input_tokens_seen": 13928436,
+      "num_tokens": 13928436.0,
+      "step": 2150
+    },
+    {
+      "epoch": 4.104477611940299,
+      "grad_norm": 1.0974746942520142,
+      "learning_rate": 7.709876238235702e-06,
+      "loss": 0.0839,
+      "mean_token_accuracy": 0.9738617813587189,
+      "num_input_tokens_seen": 14249072,
+      "num_tokens": 14249072.0,
+      "step": 2200
+    },
+    {
+      "epoch": 4.197761194029851,
+      "grad_norm": 1.0228750705718994,
+      "learning_rate": 7.29172730329028e-06,
+      "loss": 0.0498,
+      "mean_token_accuracy": 0.9851219677925109,
+      "num_input_tokens_seen": 14571152,
+      "num_tokens": 14571152.0,
+      "step": 2250
+    },
+    {
+      "epoch": 4.291044776119403,
+      "grad_norm": 0.6385033130645752,
+      "learning_rate": 6.8786293760869695e-06,
+      "loss": 0.0388,
+      "mean_token_accuracy": 0.9884346830844879,
+      "num_input_tokens_seen": 14903168,
+      "num_tokens": 14903168.0,
+      "step": 2300
+    },
+    {
+      "epoch": 4.384328358208955,
+      "grad_norm": 0.8694852590560913,
+      "learning_rate": 6.4713528963348506e-06,
+      "loss": 0.0411,
+      "mean_token_accuracy": 0.9874970281124115,
+      "num_input_tokens_seen": 15233224,
+      "num_tokens": 15233224.0,
+      "step": 2350
+    },
+    {
+      "epoch": 4.477611940298507,
+      "grad_norm": 0.929122805595398,
+      "learning_rate": 6.070657446573347e-06,
+      "loss": 0.0476,
+      "mean_token_accuracy": 0.985052285194397,
+      "num_input_tokens_seen": 15565618,
+      "num_tokens": 15565618.0,
+      "step": 2400
+    },
+    {
+      "epoch": 4.57089552238806,
+      "grad_norm": 0.6402145624160767,
+      "learning_rate": 5.677290335528576e-06,
+      "loss": 0.0694,
+      "mean_token_accuracy": 0.9864287662506104,
+      "num_input_tokens_seen": 15886728,
+      "num_tokens": 15886728.0,
+      "step": 2450
+    },
+    {
+      "epoch": 4.664179104477612,
+      "grad_norm": 0.9010692238807678,
+      "learning_rate": 5.291985204360754e-06,
+      "loss": 0.0424,
+      "mean_token_accuracy": 0.9870220470428467,
+      "num_input_tokens_seen": 16200844,
+      "num_tokens": 16200844.0,
+      "step": 2500
+    },
+    {
+      "epoch": 4.757462686567164,
+      "grad_norm": 0.8838453888893127,
+      "learning_rate": 4.9154606584019646e-06,
+      "loss": 0.0433,
+      "mean_token_accuracy": 0.9861760056018829,
+      "num_input_tokens_seen": 16518546,
+      "num_tokens": 16518546.0,
+      "step": 2550
+    },
+    {
+      "epoch": 4.850746268656716,
+      "grad_norm": 1.0508785247802734,
+      "learning_rate": 4.548418926936235e-06,
+      "loss": 0.0413,
+      "mean_token_accuracy": 0.9870886874198913,
+      "num_input_tokens_seen": 16837904,
+      "num_tokens": 16837904.0,
+      "step": 2600
+    },
+    {
+      "epoch": 4.9440298507462686,
+      "grad_norm": 1.0464740991592407,
+      "learning_rate": 4.191544553521355e-06,
+      "loss": 0.0428,
+      "mean_token_accuracy": 0.9860042917728424,
+      "num_input_tokens_seen": 17162930,
+      "num_tokens": 17162930.0,
+      "step": 2650
+    },
+    {
+      "epoch": 5.037313432835821,
+      "grad_norm": 1.167823076248169,
+      "learning_rate": 3.845503119295182e-06,
+      "loss": 0.0407,
+      "mean_token_accuracy": 0.9872146189212799,
+      "num_input_tokens_seen": 17492580,
+      "num_tokens": 17492580.0,
+      "step": 2700
+    },
+    {
+      "epoch": 5.130597014925373,
+      "grad_norm": 0.6037238836288452,
+      "learning_rate": 3.5109400016473338e-06,
+      "loss": 0.0259,
+      "mean_token_accuracy": 0.9919403278827668,
+      "num_input_tokens_seen": 17814314,
+      "num_tokens": 17814314.0,
+      "step": 2750
+    },
+    {
+      "epoch": 5.223880597014926,
+      "grad_norm": 0.4187396168708801,
+      "learning_rate": 3.1884791705714936e-06,
+      "loss": 0.0157,
+      "mean_token_accuracy": 0.9955093479156494,
+      "num_input_tokens_seen": 18139326,
+      "num_tokens": 18139326.0,
+      "step": 2800
+    },
+    {
+      "epoch": 5.317164179104478,
+      "grad_norm": 0.46902546286582947,
+      "learning_rate": 2.878722024943139e-06,
+      "loss": 0.0139,
+      "mean_token_accuracy": 0.995677514076233,
+      "num_input_tokens_seen": 18466486,
+      "num_tokens": 18466486.0,
+      "step": 2850
+    },
+    {
+      "epoch": 5.41044776119403,
+      "grad_norm": 0.3824739456176758,
+      "learning_rate": 2.5822462708930607e-06,
+      "loss": 0.0146,
+      "mean_token_accuracy": 0.9957832169532775,
+      "num_input_tokens_seen": 18796326,
+      "num_tokens": 18796326.0,
+      "step": 2900
+    },
+    {
+      "epoch": 5.503731343283582,
+      "grad_norm": 0.5814462304115295,
+      "learning_rate": 2.299604844368547e-06,
+      "loss": 0.0156,
+      "mean_token_accuracy": 0.9954328262805938,
+      "num_input_tokens_seen": 19129954,
+      "num_tokens": 19129954.0,
+      "step": 2950
+    },
+    {
+      "epoch": 5.597014925373134,
+      "grad_norm": 0.45167264342308044,
+      "learning_rate": 2.031324879891664e-06,
+      "loss": 0.013,
+      "mean_token_accuracy": 0.9963365316390991,
+      "num_input_tokens_seen": 19448322,
+      "num_tokens": 19448322.0,
+      "step": 3000
+    },
+    {
+      "epoch": 5.690298507462686,
+      "grad_norm": 0.5698373913764954,
+      "learning_rate": 1.777906727437979e-06,
+      "loss": 0.0127,
+      "mean_token_accuracy": 0.9959760665893554,
+      "num_input_tokens_seen": 19765536,
+      "num_tokens": 19765536.0,
+      "step": 3050
+    },
+    {
+      "epoch": 5.7835820895522385,
+      "grad_norm": 0.5558890104293823,
+      "learning_rate": 1.5398230192692275e-06,
+      "loss": 0.0117,
+      "mean_token_accuracy": 0.9963926291465759,
+      "num_input_tokens_seen": 20082350,
+      "num_tokens": 20082350.0,
+      "step": 3100
+    },
+    {
+      "epoch": 5.8768656716417915,
+      "grad_norm": 0.46126776933670044,
+      "learning_rate": 1.3175177884603252e-06,
+      "loss": 0.0112,
+      "mean_token_accuracy": 0.9968423485755921,
+      "num_input_tokens_seen": 20401204,
+      "num_tokens": 20401204.0,
+      "step": 3150
+    },
+    {
+      "epoch": 5.970149253731344,
+      "grad_norm": 0.4649079144001007,
+      "learning_rate": 1.1114056407647045e-06,
+      "loss": 0.0111,
+      "mean_token_accuracy": 0.9965043890476227,
+      "num_input_tokens_seen": 20730370,
+      "num_tokens": 20730370.0,
+      "step": 3200
+    },
+    {
+      "epoch": 6.063432835820896,
+      "grad_norm": 0.49266737699508667,
+      "learning_rate": 9.218709813624749e-07,
+      "loss": 0.012,
+      "mean_token_accuracy": 0.9967284095287323,
+      "num_input_tokens_seen": 21052760,
+      "num_tokens": 21052760.0,
+      "step": 3250
+    },
+    {
+      "epoch": 6.156716417910448,
+      "grad_norm": 0.38067445158958435,
+      "learning_rate": 7.492672979335147e-07,
+      "loss": 0.0069,
+      "mean_token_accuracy": 0.998069132566452,
+      "num_input_tokens_seen": 21373624,
+      "num_tokens": 21373624.0,
+      "step": 3300
+    },
+    {
+      "epoch": 6.25,
+      "grad_norm": 0.43708378076553345,
+      "learning_rate": 5.939165013926195e-07,
+      "loss": 0.0054,
+      "mean_token_accuracy": 0.9985744786262513,
+      "num_input_tokens_seen": 21701456,
+      "num_tokens": 21701456.0,
+      "step": 3350
+    },
+    {
+      "epoch": 6.343283582089552,
+      "grad_norm": 0.5964256525039673,
+      "learning_rate": 4.56108325516238e-07,
+      "loss": 0.0066,
+      "mean_token_accuracy": 0.9981687092781066,
+      "num_input_tokens_seen": 22028186,
+      "num_tokens": 22028186.0,
+      "step": 3400
+    },
+    {
+      "epoch": 6.436567164179104,
+      "grad_norm": 0.5558670163154602,
+      "learning_rate": 3.3609978658051045e-07,
+      "loss": 0.0065,
+      "mean_token_accuracy": 0.9981926000118255,
+      "num_input_tokens_seen": 22357840,
+      "num_tokens": 22357840.0,
+      "step": 3450
+    },
+    {
+      "epoch": 6.529850746268656,
+      "grad_norm": 0.5026367902755737,
+      "learning_rate": 2.341147040184011e-07,
+      "loss": 0.0052,
+      "mean_token_accuracy": 0.9985788524150848,
+      "num_input_tokens_seen": 22688738,
+      "num_tokens": 22688738.0,
+      "step": 3500
+    },
+    {
+      "epoch": 6.6231343283582085,
+      "grad_norm": 0.2897884249687195,
+      "learning_rate": 1.5034328298990652e-07,
+      "loss": 0.0046,
+      "mean_token_accuracy": 0.9987975347042084,
+      "num_input_tokens_seen": 23010742,
+      "num_tokens": 23010742.0,
+      "step": 3550
+    },
+    {
+      "epoch": 6.7164179104477615,
+      "grad_norm": 0.43537065386772156,
+      "learning_rate": 8.494175964388285e-08,
+      "loss": 0.0039,
+      "mean_token_accuracy": 0.9988950288295746,
+      "num_input_tokens_seen": 23329282,
+      "num_tokens": 23329282.0,
+      "step": 3600
+    },
+    {
+      "epoch": 6.809701492537314,
+      "grad_norm": 0.36725738644599915,
+      "learning_rate": 3.803210973305715e-08,
+      "loss": 0.003,
+      "mean_token_accuracy": 0.9991644871234894,
+      "num_input_tokens_seen": 23647522,
+      "num_tokens": 23647522.0,
+      "step": 3650
+    },
+    {
+      "epoch": 6.902985074626866,
+      "grad_norm": 0.531343400478363,
+      "learning_rate": 9.7018211256783e-09,
+      "loss": 0.0046,
+      "mean_token_accuracy": 0.9987628531455993,
+      "num_input_tokens_seen": 23963558,
+      "num_tokens": 23963558.0,
+      "step": 3700
+    },
+    {
+      "epoch": 6.996268656716418,
+      "grad_norm": 0.6237491369247437,
+      "learning_rate": 3.7306380940016486e-12,
+      "loss": 0.0051,
+      "mean_token_accuracy": 0.9986482226848602,
+      "num_input_tokens_seen": 24290780,
+      "num_tokens": 24290780.0,
+      "step": 3750
+    },
+    {
+      "epoch": 6.996268656716418,
+      "num_input_tokens_seen": 24290780,
+      "step": 3750,
+      "total_flos": 1.4113858770069094e+17,
+      "train_loss": 0.21003443336486816,
+      "train_runtime": 66532.5206,
+      "train_samples_per_second": 0.451,
+      "train_steps_per_second": 0.056,
+      "train_tokens_per_second": 70.398
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 3750,
+  "num_input_tokens_seen": 24290780,
+  "num_train_epochs": 7,
+  "save_steps": 0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.4113858770069094e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ceac200286f0cf175e77ab1763672204503403d9d5009bfa55db268c97ee492
+size 6161