Upload 20 files

.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

README.md

@@ -0,0 +1,58 @@
+---
+library_name: peft
+license: other
+base_model: Qwen/Qwen2.5-VL-7B-Instruct
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: BS_riche_lora
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# BS_riche_lora
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct) on the BsKIE3 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 16
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- num_epochs: 3.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.49.0
+- Pytorch 2.4.0+cu121
+- Datasets 2.21.0
+- Tokenizers 0.21.0

adapter_config.json

@@ -0,0 +1,223 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "model.layers.17.self_attn.k_proj",
+    "model.layers.16.mlp.gate_proj",
+    "model.layers.18.self_attn.k_proj",
+    "model.layers.24.mlp.up_proj",
+    "model.layers.20.mlp.gate_proj",
+    "model.layers.5.mlp.gate_proj",
+    "model.layers.20.self_attn.k_proj",
+    "model.layers.15.mlp.gate_proj",
+    "model.layers.21.self_attn.o_proj",
+    "model.layers.17.mlp.up_proj",
+    "model.layers.12.self_attn.q_proj",
+    "model.layers.4.mlp.up_proj",
+    "model.layers.12.mlp.gate_proj",
+    "model.layers.6.mlp.down_proj",
+    "model.layers.7.mlp.up_proj",
+    "model.layers.6.self_attn.k_proj",
+    "model.layers.19.self_attn.q_proj",
+    "model.layers.20.mlp.down_proj",
+    "model.layers.5.self_attn.q_proj",
+    "model.layers.25.mlp.down_proj",
+    "model.layers.10.mlp.gate_proj",
+    "model.layers.9.mlp.up_proj",
+    "model.layers.13.self_attn.q_proj",
+    "model.layers.21.mlp.gate_proj",
+    "model.layers.0.mlp.down_proj",
+    "model.layers.10.self_attn.v_proj",
+    "model.layers.18.self_attn.v_proj",
+    "model.layers.11.self_attn.v_proj",
+    "model.layers.0.self_attn.q_proj",
+    "model.layers.1.mlp.gate_proj",
+    "model.layers.17.self_attn.o_proj",
+    "model.layers.9.self_attn.o_proj",
+    "model.layers.22.mlp.gate_proj",
+    "model.layers.10.self_attn.q_proj",
+    "model.layers.26.mlp.down_proj",
+    "model.layers.1.self_attn.o_proj",
+    "model.layers.19.mlp.up_proj",
+    "model.layers.7.self_attn.v_proj",
+    "model.layers.22.mlp.up_proj",
+    "model.layers.17.self_attn.q_proj",
+    "model.layers.3.mlp.gate_proj",
+    "model.layers.1.self_attn.v_proj",
+    "model.layers.13.self_attn.v_proj",
+    "model.layers.0.mlp.up_proj",
+    "model.layers.3.self_attn.v_proj",
+    "model.layers.5.mlp.down_proj",
+    "model.layers.4.self_attn.k_proj",
+    "model.layers.4.self_attn.o_proj",
+    "model.layers.10.self_attn.k_proj",
+    "model.layers.14.self_attn.v_proj",
+    "model.layers.16.self_attn.q_proj",
+    "model.layers.4.mlp.down_proj",
+    "model.layers.6.mlp.gate_proj",
+    "model.layers.9.mlp.down_proj",
+    "model.layers.18.mlp.down_proj",
+    "model.layers.26.self_attn.o_proj",
+    "model.layers.1.self_attn.k_proj",
+    "model.layers.14.mlp.down_proj",
+    "model.layers.1.self_attn.q_proj",
+    "model.layers.8.self_attn.o_proj",
+    "model.layers.23.self_attn.o_proj",
+    "model.layers.11.self_attn.o_proj",
+    "model.layers.21.self_attn.v_proj",
+    "model.layers.7.self_attn.q_proj",
+    "model.layers.11.self_attn.k_proj",
+    "model.layers.12.self_attn.o_proj",
+    "model.layers.20.self_attn.o_proj",
+    "model.layers.7.self_attn.k_proj",
+    "model.layers.7.self_attn.o_proj",
+    "model.layers.16.self_attn.o_proj",
+    "model.layers.8.self_attn.v_proj",
+    "model.layers.17.mlp.gate_proj",
+    "model.layers.3.self_attn.o_proj",
+    "model.layers.11.mlp.up_proj",
+    "model.layers.23.mlp.down_proj",
+    "model.layers.27.self_attn.k_proj",
+    "model.layers.27.mlp.up_proj",
+    "model.layers.8.mlp.down_proj",
+    "model.layers.7.mlp.gate_proj",
+    "model.layers.5.self_attn.k_proj",
+    "model.layers.20.self_attn.v_proj",
+    "model.layers.25.self_attn.k_proj",
+    "model.layers.10.self_attn.o_proj",
+    "model.layers.22.self_attn.o_proj",
+    "model.layers.24.mlp.gate_proj",
+    "model.layers.7.mlp.down_proj",
+    "model.layers.14.self_attn.k_proj",
+    "model.layers.16.self_attn.k_proj",
+    "model.layers.24.self_attn.o_proj",
+    "model.layers.6.self_attn.q_proj",
+    "model.layers.4.self_attn.v_proj",
+    "model.layers.14.self_attn.o_proj",
+    "model.layers.0.self_attn.o_proj",
+    "model.layers.27.self_attn.v_proj",
+    "model.layers.18.self_attn.o_proj",
+    "model.layers.27.self_attn.q_proj",
+    "model.layers.8.mlp.up_proj",
+    "model.layers.25.mlp.gate_proj",
+    "model.layers.6.self_attn.v_proj",
+    "model.layers.9.mlp.gate_proj",
+    "model.layers.15.self_attn.q_proj",
+    "model.layers.8.mlp.gate_proj",
+    "model.layers.13.mlp.up_proj",
+    "model.layers.4.mlp.gate_proj",
+    "model.layers.19.mlp.down_proj",
+    "model.layers.19.self_attn.k_proj",
+    "model.layers.23.mlp.gate_proj",
+    "model.layers.22.self_attn.q_proj",
+    "model.layers.13.self_attn.k_proj",
+    "model.layers.15.self_attn.o_proj",
+    "model.layers.26.self_attn.v_proj",
+    "model.layers.1.mlp.down_proj",
+    "model.layers.14.mlp.gate_proj",
+    "model.layers.12.mlp.up_proj",
+    "model.layers.26.self_attn.q_proj",
+    "model.layers.14.self_attn.q_proj",
+    "model.layers.3.mlp.down_proj",
+    "model.layers.24.self_attn.k_proj",
+    "model.layers.3.self_attn.k_proj",
+    "model.layers.20.mlp.up_proj",
+    "model.layers.21.mlp.down_proj",
+    "model.layers.10.mlp.down_proj",
+    "model.layers.27.mlp.down_proj",
+    "model.layers.26.mlp.gate_proj",
+    "model.layers.16.mlp.up_proj",
+    "model.layers.0.self_attn.k_proj",
+    "model.layers.21.mlp.up_proj",
+    "model.layers.6.self_attn.o_proj",
+    "model.layers.27.mlp.gate_proj",
+    "model.layers.22.mlp.down_proj",
+    "model.layers.3.self_attn.q_proj",
+    "model.layers.15.mlp.up_proj",
+    "model.layers.8.self_attn.q_proj",
+    "model.layers.17.self_attn.v_proj",
+    "model.layers.20.self_attn.q_proj",
+    "model.layers.12.self_attn.v_proj",
+    "model.layers.16.mlp.down_proj",
+    "model.layers.2.self_attn.o_proj",
+    "model.layers.19.self_attn.v_proj",
+    "model.layers.22.self_attn.k_proj",
+    "model.layers.10.mlp.up_proj",
+    "model.layers.2.mlp.gate_proj",
+    "model.layers.2.self_attn.k_proj",
+    "model.layers.9.self_attn.q_proj",
+    "model.layers.25.self_attn.q_proj",
+    "model.layers.18.mlp.up_proj",
+    "model.layers.25.mlp.up_proj",
+    "model.layers.1.mlp.up_proj",
+    "model.layers.5.mlp.up_proj",
+    "model.layers.2.self_attn.v_proj",
+    "model.layers.13.mlp.gate_proj",
+    "model.layers.18.mlp.gate_proj",
+    "model.layers.11.mlp.gate_proj",
+    "model.layers.19.self_attn.o_proj",
+    "model.layers.11.mlp.down_proj",
+    "model.layers.18.self_attn.q_proj",
+    "model.layers.17.mlp.down_proj",
+    "model.layers.2.mlp.down_proj",
+    "model.layers.13.mlp.down_proj",
+    "model.layers.16.self_attn.v_proj",
+    "model.layers.15.self_attn.k_proj",
+    "model.layers.15.self_attn.v_proj",
+    "model.layers.3.mlp.up_proj",
+    "model.layers.15.mlp.down_proj",
+    "model.layers.12.mlp.down_proj",
+    "model.layers.13.self_attn.o_proj",
+    "model.layers.23.mlp.up_proj",
+    "model.layers.22.self_attn.v_proj",
+    "model.layers.21.self_attn.k_proj",
+    "model.layers.26.mlp.up_proj",
+    "model.layers.26.self_attn.k_proj",
+    "model.layers.2.mlp.up_proj",
+    "model.layers.23.self_attn.q_proj",
+    "model.layers.19.mlp.gate_proj",
+    "model.layers.24.self_attn.v_proj",
+    "model.layers.4.self_attn.q_proj",
+    "model.layers.9.self_attn.v_proj",
+    "model.layers.14.mlp.up_proj",
+    "model.layers.25.self_attn.v_proj",
+    "model.layers.12.self_attn.k_proj",
+    "model.layers.24.mlp.down_proj",
+    "model.layers.27.self_attn.o_proj",
+    "model.layers.24.self_attn.q_proj",
+    "model.layers.5.self_attn.o_proj",
+    "model.layers.6.mlp.up_proj",
+    "model.layers.0.mlp.gate_proj",
+    "model.layers.5.self_attn.v_proj",
+    "model.layers.23.self_attn.v_proj",
+    "model.layers.9.self_attn.k_proj",
+    "model.layers.25.self_attn.o_proj",
+    "model.layers.2.self_attn.q_proj",
+    "model.layers.8.self_attn.k_proj",
+    "model.layers.21.self_attn.q_proj",
+    "model.layers.23.self_attn.k_proj",
+    "model.layers.11.self_attn.q_proj",
+    "model.layers.0.self_attn.v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3a73b0aff9828aeea7d13edf6709ee37a47343efdd235fa78d90afcc31696cf
+size 80792096

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
+}

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.9904761904761905,
+    "num_input_tokens_seen": 8039616,
+    "total_flos": 3.746792787278561e+17,
+    "train_loss": 0.06834620604148278,
+    "train_runtime": 3970.1665,
+    "train_samples_per_second": 0.793,
+    "train_steps_per_second": 0.049
+}

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'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

llamaboard_config.yaml

@@ -0,0 +1,78 @@
+top.booster: auto
+top.checkpoint_path: []
+top.finetuning_type: lora
+top.model_name: Qwen2.5-VL-7B-Instruct
+top.quantization_bit: none
+top.quantization_method: bitsandbytes
+top.rope_scaling: none
+top.template: qwen2_vl
+train.additional_target: ''
+train.apollo_rank: 16
+train.apollo_scale: 32
+train.apollo_target: all
+train.apollo_update_interval: 200
+train.badam_mode: layer
+train.badam_switch_interval: 50
+train.badam_switch_mode: ascending
+train.badam_update_ratio: 0.05
+train.batch_size: 2
+train.compute_type: bf16
+train.create_new_adapter: false
+train.cutoff_len: 34114
+train.dataset:
+- BsKIE3
+train.dataset_dir: data
+train.ds_offload: false
+train.ds_stage: none
+train.extra_args: '{"optim": "adamw_torch"}'
+train.freeze_extra_modules: ''
+train.freeze_trainable_layers: 2
+train.freeze_trainable_modules: all
+train.galore_rank: 16
+train.galore_scale: 2
+train.galore_target: all
+train.galore_update_interval: 200
+train.gradient_accumulation_steps: 8
+train.learning_rate: 5e-5
+train.logging_steps: 5
+train.lora_alpha: 16
+train.lora_dropout: 0
+train.lora_rank: 8
+train.lora_target: ''
+train.loraplus_lr_ratio: 0
+train.lr_scheduler_type: cosine
+train.mask_history: false
+train.max_grad_norm: '1.0'
+train.max_samples: '100000'
+train.neat_packing: false
+train.neftune_alpha: 0
+train.num_train_epochs: '3.0'
+train.packing: false
+train.ppo_score_norm: false
+train.ppo_whiten_rewards: false
+train.pref_beta: 0.1
+train.pref_ftx: 0
+train.pref_loss: sigmoid
+train.report_to:
+- none
+train.resize_vocab: false
+train.reward_model: null
+train.save_steps: 100
+train.swanlab_api_key: ''
+train.swanlab_link: ''
+train.swanlab_mode: cloud
+train.swanlab_project: llamafactory
+train.swanlab_run_name: ''
+train.swanlab_workspace: ''
+train.train_on_prompt: false
+train.training_stage: Supervised Fine-Tuning
+train.use_apollo: false
+train.use_badam: false
+train.use_dora: false
+train.use_galore: false
+train.use_llama_pro: false
+train.use_pissa: false
+train.use_rslora: false
+train.use_swanlab: false
+train.val_size: 0
+train.warmup_steps: 0

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
+}

running_log.txt

@@ -0,0 +1,601 @@
+[INFO|2025-03-21 07:37:06] configuration_utils.py:699 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/config.json
+
+[INFO|2025-03-21 07:37:06] configuration_utils.py:771 >> Model config Qwen2_5_VLConfig {
+  "_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
+  "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": "bfloat16",
+  "transformers_version": "4.49.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "hidden_size": 1280,
+    "in_chans": 3,
+    "model_type": "qwen2_5_vl",
+    "spatial_patch_size": 14,
+    "tokens_per_second": 2
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-03-21 07:37:06] tokenization_utils_base.py:2050 >> loading file vocab.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/vocab.json
+
+[INFO|2025-03-21 07:37:06] tokenization_utils_base.py:2050 >> loading file merges.txt from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/merges.txt
+
+[INFO|2025-03-21 07:37:06] tokenization_utils_base.py:2050 >> loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/tokenizer.json
+
+[INFO|2025-03-21 07:37:06] tokenization_utils_base.py:2050 >> loading file added_tokens.json from cache at None
+
+[INFO|2025-03-21 07:37:06] tokenization_utils_base.py:2050 >> loading file special_tokens_map.json from cache at None
+
+[INFO|2025-03-21 07:37:06] tokenization_utils_base.py:2050 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/tokenizer_config.json
+
+[INFO|2025-03-21 07:37:06] tokenization_utils_base.py:2050 >> loading file chat_template.jinja from cache at None
+
+[INFO|2025-03-21 07:37:06] tokenization_utils_base.py:2313 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+[INFO|2025-03-21 07:37:07] image_processing_base.py:381 >> loading configuration file preprocessor_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/preprocessor_config.json
+
+[INFO|2025-03-21 07:37:07] image_processing_base.py:381 >> loading configuration file preprocessor_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/preprocessor_config.json
+
+[WARNING|2025-03-21 07:37:07] logging.py:329 >> Using a slow image processor as `use_fast` is unset and a slow processor was saved with this model. `use_fast=True` will be the default behavior in v4.48, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with `use_fast=False`.
+
+[INFO|2025-03-21 07:37:07] image_processing_base.py:434 >> Image processor Qwen2VLImageProcessor {
+  "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
+}
+
+
+[INFO|2025-03-21 07:37:07] tokenization_utils_base.py:2050 >> loading file vocab.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/vocab.json
+
+[INFO|2025-03-21 07:37:07] tokenization_utils_base.py:2050 >> loading file merges.txt from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/merges.txt
+
+[INFO|2025-03-21 07:37:07] tokenization_utils_base.py:2050 >> loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/tokenizer.json
+
+[INFO|2025-03-21 07:37:07] tokenization_utils_base.py:2050 >> loading file added_tokens.json from cache at None
+
+[INFO|2025-03-21 07:37:07] tokenization_utils_base.py:2050 >> loading file special_tokens_map.json from cache at None
+
+[INFO|2025-03-21 07:37:07] tokenization_utils_base.py:2050 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/tokenizer_config.json
+
+[INFO|2025-03-21 07:37:07] tokenization_utils_base.py:2050 >> loading file chat_template.jinja from cache at None
+
+[INFO|2025-03-21 07:37:07] tokenization_utils_base.py:2313 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+[INFO|2025-03-21 07:37:08] processing_utils.py:876 >> Processor Qwen2_5_VLProcessor:
+- image_processor: Qwen2VLImageProcessor {
+  "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
+}
+
+- tokenizer: Qwen2TokenizerFast(name_or_path='Qwen/Qwen2.5-VL-7B-Instruct', vocab_size=151643, model_max_length=131072, is_fast=True, padding_side='right', truncation_side='right', special_tokens={'eos_token': '<|im_end|>', 'pad_token': '<|endoftext|>', '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|>']}, clean_up_tokenization_spaces=False, added_tokens_decoder={
+	151643: AddedToken("<|endoftext|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151644: AddedToken("<|im_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151645: AddedToken("<|im_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151646: AddedToken("<|object_ref_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151647: AddedToken("<|object_ref_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151648: AddedToken("<|box_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151649: AddedToken("<|box_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151650: AddedToken("<|quad_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151651: AddedToken("<|quad_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151652: AddedToken("<|vision_start|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151653: AddedToken("<|vision_end|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151654: AddedToken("<|vision_pad|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151655: AddedToken("<|image_pad|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151656: AddedToken("<|video_pad|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	151657: AddedToken("<tool_call>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),
+	151658: AddedToken("</tool_call>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),
+	151659: AddedToken("<|fim_prefix|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),
+	151660: AddedToken("<|fim_middle|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),
+	151661: AddedToken("<|fim_suffix|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),
+	151662: AddedToken("<|fim_pad|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),
+	151663: AddedToken("<|repo_name|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),
+	151664: AddedToken("<|file_sep|>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),
+}
+)
+
+{
+  "processor_class": "Qwen2_5_VLProcessor"
+}
+
+
+[INFO|2025-03-21 07:37:08] logging.py:157 >> Add <|im_end|> to stop words.
+
+[INFO|2025-03-21 07:37:08] logging.py:157 >> Loading dataset BsKIE3.json...
+
+[INFO|2025-03-21 07:37:21] configuration_utils.py:699 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/config.json
+
+[INFO|2025-03-21 07:37:21] configuration_utils.py:771 >> Model config Qwen2_5_VLConfig {
+  "_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
+  "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": "bfloat16",
+  "transformers_version": "4.49.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "hidden_size": 1280,
+    "in_chans": 3,
+    "model_type": "qwen2_5_vl",
+    "spatial_patch_size": 14,
+    "tokens_per_second": 2
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-03-21 07:37:21] modeling_utils.py:3982 >> loading weights file model.safetensors from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/model.safetensors.index.json
+
+[INFO|2025-03-21 07:37:21] modeling_utils.py:1633 >> Instantiating Qwen2_5_VLForConditionalGeneration model under default dtype torch.bfloat16.
+
+[INFO|2025-03-21 07:37:21] configuration_utils.py:1140 >> Generate config GenerationConfig {
+  "bos_token_id": 151643,
+  "eos_token_id": 151645
+}
+
+
+[INFO|2025-03-21 07:37:21] modeling_utils.py:1633 >> Instantiating Qwen2_5_VisionTransformerPretrainedModel model under default dtype torch.bfloat16.
+
+[INFO|2025-03-21 07:37:25] modeling_utils.py:4970 >> All model checkpoint weights were used when initializing Qwen2_5_VLForConditionalGeneration.
+
+
+[INFO|2025-03-21 07:37:25] modeling_utils.py:4978 >> All the weights of Qwen2_5_VLForConditionalGeneration were initialized from the model checkpoint at Qwen/Qwen2.5-VL-7B-Instruct.
+If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2_5_VLForConditionalGeneration for predictions without further training.
+
+[INFO|2025-03-21 07:37:25] configuration_utils.py:1095 >> loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/generation_config.json
+
+[INFO|2025-03-21 07:37:25] configuration_utils.py:1140 >> Generate config GenerationConfig {
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.1,
+  "top_k": 1,
+  "top_p": 0.001
+}
+
+
+[INFO|2025-03-21 07:37:26] logging.py:157 >> Gradient checkpointing enabled.
+
+[INFO|2025-03-21 07:37:26] logging.py:157 >> Using torch SDPA for faster training and inference.
+
+[INFO|2025-03-21 07:37:26] logging.py:157 >> Upcasting trainable params to float32.
+
+[INFO|2025-03-21 07:37:26] logging.py:157 >> Fine-tuning method: LoRA
+
+[INFO|2025-03-21 07:37:26] logging.py:157 >> Found linear modules: gate_proj,up_proj,v_proj,down_proj,q_proj,k_proj,o_proj
+
+[INFO|2025-03-21 07:37:26] logging.py:157 >> Set vision model not trainable: ['visual.patch_embed', 'visual.blocks'].
+
+[INFO|2025-03-21 07:37:26] logging.py:157 >> Set multi model projector not trainable: visual.merger.
+
+[INFO|2025-03-21 07:37:26] logging.py:157 >> trainable params: 20,185,088 || all params: 8,312,351,744 || trainable%: 0.2428
+
+[INFO|2025-03-21 07:37:26] trainer.py:746 >> Using auto half precision backend
+
+[WARNING|2025-03-21 07:37:26] trainer.py:781 >> No label_names provided for model class `PeftModelForCausalLM`. Since `PeftModel` hides base models input arguments, if label_names is not given, label_names can't be set automatically within `Trainer`. Note that empty label_names list will be used instead.
+
+[INFO|2025-03-21 07:37:26] trainer.py:2405 >> ***** Running training *****
+
+[INFO|2025-03-21 07:37:26] trainer.py:2406 >>   Num examples = 1,050
+
+[INFO|2025-03-21 07:37:26] trainer.py:2407 >>   Num Epochs = 3
+
+[INFO|2025-03-21 07:37:26] trainer.py:2408 >>   Instantaneous batch size per device = 2
+
+[INFO|2025-03-21 07:37:26] trainer.py:2411 >>   Total train batch size (w. parallel, distributed & accumulation) = 16
+
+[INFO|2025-03-21 07:37:26] trainer.py:2412 >>   Gradient Accumulation steps = 8
+
+[INFO|2025-03-21 07:37:26] trainer.py:2413 >>   Total optimization steps = 195
+
+[INFO|2025-03-21 07:37:26] trainer.py:2414 >>   Number of trainable parameters = 20,185,088
+
+[INFO|2025-03-21 07:39:08] logging.py:157 >> {'loss': 0.3154, 'learning_rate': 4.9919e-05, 'epoch': 0.08, 'throughput': 2074.94}
+
+[INFO|2025-03-21 07:40:50] logging.py:157 >> {'loss': 0.2409, 'learning_rate': 4.9676e-05, 'epoch': 0.15, 'throughput': 2059.35}
+
+[INFO|2025-03-21 07:42:36] logging.py:157 >> {'loss': 0.1711, 'learning_rate': 4.9274e-05, 'epoch': 0.23, 'throughput': 2038.91}
+
+[INFO|2025-03-21 07:44:18] logging.py:157 >> {'loss': 0.1228, 'learning_rate': 4.8713e-05, 'epoch': 0.30, 'throughput': 2033.99}
+
+[INFO|2025-03-21 07:45:59] logging.py:157 >> {'loss': 0.0945, 'learning_rate': 4.7999e-05, 'epoch': 0.38, 'throughput': 2030.92}
+
+[INFO|2025-03-21 07:47:40] logging.py:157 >> {'loss': 0.0873, 'learning_rate': 4.7136e-05, 'epoch': 0.46, 'throughput': 2028.08}
+
+[INFO|2025-03-21 07:49:18] logging.py:157 >> {'loss': 0.0635, 'learning_rate': 4.6130e-05, 'epoch': 0.53, 'throughput': 2030.96}
+
+[INFO|2025-03-21 07:51:00] logging.py:157 >> {'loss': 0.0784, 'learning_rate': 4.4986e-05, 'epoch': 0.61, 'throughput': 2029.27}
+
+[INFO|2025-03-21 07:52:36] logging.py:157 >> {'loss': 0.0634, 'learning_rate': 4.3713e-05, 'epoch': 0.69, 'throughput': 2032.57}
+
+[INFO|2025-03-21 07:54:19] logging.py:157 >> {'loss': 0.0715, 'learning_rate': 4.2318e-05, 'epoch': 0.76, 'throughput': 2030.24}
+
+[INFO|2025-03-21 07:56:02] logging.py:157 >> {'loss': 0.0588, 'learning_rate': 4.0811e-05, 'epoch': 0.84, 'throughput': 2030.66}
+
+[INFO|2025-03-21 07:57:43] logging.py:157 >> {'loss': 0.0634, 'learning_rate': 3.9202e-05, 'epoch': 0.91, 'throughput': 2030.21}
+
+[INFO|2025-03-21 07:59:24] logging.py:157 >> {'loss': 0.0582, 'learning_rate': 3.7500e-05, 'epoch': 0.99, 'throughput': 2030.85}
+
+[INFO|2025-03-21 08:01:14] logging.py:157 >> {'loss': 0.0687, 'learning_rate': 3.5717e-05, 'epoch': 1.08, 'throughput': 2029.43}
+
+[INFO|2025-03-21 08:02:45] logging.py:157 >> {'loss': 0.0539, 'learning_rate': 3.3865e-05, 'epoch': 1.15, 'throughput': 2033.85}
+
+[INFO|2025-03-21 08:04:29] logging.py:157 >> {'loss': 0.0545, 'learning_rate': 3.1955e-05, 'epoch': 1.23, 'throughput': 2030.86}
+
+[INFO|2025-03-21 08:06:14] logging.py:157 >> {'loss': 0.0446, 'learning_rate': 3.0001e-05, 'epoch': 1.30, 'throughput': 2030.02}
+
+[INFO|2025-03-21 08:07:52] logging.py:157 >> {'loss': 0.0535, 'learning_rate': 2.8013e-05, 'epoch': 1.38, 'throughput': 2030.47}
+
+[INFO|2025-03-21 08:09:35] logging.py:157 >> {'loss': 0.0488, 'learning_rate': 2.6007e-05, 'epoch': 1.46, 'throughput': 2028.22}
+
+[INFO|2025-03-21 08:11:20] logging.py:157 >> {'loss': 0.0440, 'learning_rate': 2.3993e-05, 'epoch': 1.53, 'throughput': 2026.11}
+
+[INFO|2025-03-21 08:11:20] trainer.py:3942 >> Saving model checkpoint to saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-100
+
+[INFO|2025-03-21 08:11:20] configuration_utils.py:699 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/config.json
+
+[INFO|2025-03-21 08:11:20] configuration_utils.py:771 >> Model config Qwen2_5_VLConfig {
+  "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": "bfloat16",
+  "transformers_version": "4.49.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "hidden_size": 1280,
+    "in_chans": 3,
+    "model_type": "qwen2_5_vl",
+    "spatial_patch_size": 14,
+    "tokens_per_second": 2
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-03-21 08:11:21] tokenization_utils_base.py:2500 >> tokenizer config file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-100/tokenizer_config.json
+
+[INFO|2025-03-21 08:11:21] tokenization_utils_base.py:2509 >> Special tokens file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-100/special_tokens_map.json
+
+[INFO|2025-03-21 08:11:21] image_processing_base.py:261 >> Image processor saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-100/preprocessor_config.json
+
+[INFO|2025-03-21 08:11:21] tokenization_utils_base.py:2500 >> tokenizer config file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-100/tokenizer_config.json
+
+[INFO|2025-03-21 08:11:21] tokenization_utils_base.py:2509 >> Special tokens file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-100/special_tokens_map.json
+
+[INFO|2025-03-21 08:11:21] processing_utils.py:638 >> chat template saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-100/chat_template.json
+
+[INFO|2025-03-21 08:13:03] logging.py:157 >> {'loss': 0.0479, 'learning_rate': 2.1987e-05, 'epoch': 1.61, 'throughput': 2024.85}
+
+[INFO|2025-03-21 08:14:49] logging.py:157 >> {'loss': 0.0441, 'learning_rate': 1.9999e-05, 'epoch': 1.69, 'throughput': 2023.58}
+
+[INFO|2025-03-21 08:16:31] logging.py:157 >> {'loss': 0.0428, 'learning_rate': 1.8045e-05, 'epoch': 1.76, 'throughput': 2023.28}
+
+[INFO|2025-03-21 08:18:15] logging.py:157 >> {'loss': 0.0425, 'learning_rate': 1.6135e-05, 'epoch': 1.84, 'throughput': 2023.49}
+
+[INFO|2025-03-21 08:19:51] logging.py:157 >> {'loss': 0.0437, 'learning_rate': 1.4283e-05, 'epoch': 1.91, 'throughput': 2025.72}
+
+[INFO|2025-03-21 08:21:30] logging.py:157 >> {'loss': 0.0390, 'learning_rate': 1.2500e-05, 'epoch': 1.99, 'throughput': 2026.78}
+
+[INFO|2025-03-21 08:23:19] logging.py:157 >> {'loss': 0.0465, 'learning_rate': 1.0798e-05, 'epoch': 2.08, 'throughput': 2026.32}
+
+[INFO|2025-03-21 08:24:54] logging.py:157 >> {'loss': 0.0451, 'learning_rate': 9.1889e-06, 'epoch': 2.15, 'throughput': 2027.94}
+
+[INFO|2025-03-21 08:26:41] logging.py:157 >> {'loss': 0.0385, 'learning_rate': 7.6819e-06, 'epoch': 2.23, 'throughput': 2026.17}
+
+[INFO|2025-03-21 08:28:17] logging.py:157 >> {'loss': 0.0506, 'learning_rate': 6.2872e-06, 'epoch': 2.30, 'throughput': 2027.23}
+
+[INFO|2025-03-21 08:29:57] logging.py:157 >> {'loss': 0.0471, 'learning_rate': 5.0139e-06, 'epoch': 2.38, 'throughput': 2027.90}
+
+[INFO|2025-03-21 08:31:37] logging.py:157 >> {'loss': 0.0383, 'learning_rate': 3.8702e-06, 'epoch': 2.46, 'throughput': 2028.25}
+
+[INFO|2025-03-21 08:33:19] logging.py:157 >> {'loss': 0.0360, 'learning_rate': 2.8636e-06, 'epoch': 2.53, 'throughput': 2028.63}
+
+[INFO|2025-03-21 08:34:59] logging.py:157 >> {'loss': 0.0381, 'learning_rate': 2.0005e-06, 'epoch': 2.61, 'throughput': 2028.93}
+
+[INFO|2025-03-21 08:36:39] logging.py:157 >> {'loss': 0.0415, 'learning_rate': 1.2866e-06, 'epoch': 2.69, 'throughput': 2028.93}
+
+[INFO|2025-03-21 08:38:24] logging.py:157 >> {'loss': 0.0397, 'learning_rate': 7.2645e-07, 'epoch': 2.76, 'throughput': 2027.85}
+
+[INFO|2025-03-21 08:40:15] logging.py:157 >> {'loss': 0.0421, 'learning_rate': 3.2374e-07, 'epoch': 2.84, 'throughput': 2025.27}
+
+[INFO|2025-03-21 08:41:56] logging.py:157 >> {'loss': 0.0378, 'learning_rate': 8.1067e-08, 'epoch': 2.91, 'throughput': 2025.73}
+
+[INFO|2025-03-21 08:43:34] logging.py:157 >> {'loss': 0.0470, 'learning_rate': 0.0000e+00, 'epoch': 2.99, 'throughput': 2026.33}
+
+[INFO|2025-03-21 08:43:34] trainer.py:3942 >> Saving model checkpoint to saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-195
+
+[INFO|2025-03-21 08:43:35] configuration_utils.py:699 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/config.json
+
+[INFO|2025-03-21 08:43:35] configuration_utils.py:771 >> Model config Qwen2_5_VLConfig {
+  "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": "bfloat16",
+  "transformers_version": "4.49.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "hidden_size": 1280,
+    "in_chans": 3,
+    "model_type": "qwen2_5_vl",
+    "spatial_patch_size": 14,
+    "tokens_per_second": 2
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-03-21 08:43:36] tokenization_utils_base.py:2500 >> tokenizer config file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-195/tokenizer_config.json
+
+[INFO|2025-03-21 08:43:36] tokenization_utils_base.py:2509 >> Special tokens file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-195/special_tokens_map.json
+
+[INFO|2025-03-21 08:43:36] image_processing_base.py:261 >> Image processor saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-195/preprocessor_config.json
+
+[INFO|2025-03-21 08:43:36] tokenization_utils_base.py:2500 >> tokenizer config file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-195/tokenizer_config.json
+
+[INFO|2025-03-21 08:43:36] tokenization_utils_base.py:2509 >> Special tokens file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-195/special_tokens_map.json
+
+[INFO|2025-03-21 08:43:36] processing_utils.py:638 >> chat template saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/checkpoint-195/chat_template.json
+
+[INFO|2025-03-21 08:43:36] trainer.py:2657 >> 
+
+Training completed. Do not forget to share your model on huggingface.co/models =)
+
+
+
+[INFO|2025-03-21 08:43:36] image_processing_base.py:261 >> Image processor saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/preprocessor_config.json
+
+[INFO|2025-03-21 08:43:36] tokenization_utils_base.py:2500 >> tokenizer config file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/tokenizer_config.json
+
+[INFO|2025-03-21 08:43:36] tokenization_utils_base.py:2509 >> Special tokens file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/special_tokens_map.json
+
+[INFO|2025-03-21 08:43:37] processing_utils.py:638 >> chat template saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/chat_template.json
+
+[INFO|2025-03-21 08:43:37] trainer.py:3942 >> Saving model checkpoint to saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora
+
+[INFO|2025-03-21 08:43:37] configuration_utils.py:699 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/68156fd997cdc9f710620466735af49862bb81f6/config.json
+
+[INFO|2025-03-21 08:43:37] configuration_utils.py:771 >> Model config Qwen2_5_VLConfig {
+  "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": "bfloat16",
+  "transformers_version": "4.49.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "hidden_size": 1280,
+    "in_chans": 3,
+    "model_type": "qwen2_5_vl",
+    "spatial_patch_size": 14,
+    "tokens_per_second": 2
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 152064
+}
+
+
+[INFO|2025-03-21 08:43:37] tokenization_utils_base.py:2500 >> tokenizer config file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/tokenizer_config.json
+
+[INFO|2025-03-21 08:43:37] tokenization_utils_base.py:2509 >> Special tokens file saved in saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora/special_tokens_map.json
+
+[WARNING|2025-03-21 08:43:37] logging.py:162 >> No metric eval_loss to plot.
+
+[WARNING|2025-03-21 08:43:37] logging.py:162 >> No metric eval_accuracy to plot.
+
+[INFO|2025-03-21 08:43:37] modelcard.py:449 >> Dropping the following result as it does not have all the necessary fields:
+{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
+

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
+  }
+}

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": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 34114,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2_5_VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.9904761904761905,
+    "num_input_tokens_seen": 8039616,
+    "total_flos": 3.746792787278561e+17,
+    "train_loss": 0.06834620604148278,
+    "train_runtime": 3970.1665,
+    "train_samples_per_second": 0.793,
+    "train_steps_per_second": 0.049
+}

trainer_log.jsonl

@@ -0,0 +1,40 @@
+{"current_steps": 5, "total_steps": 195, "loss": 0.3154, "lr": 4.9918932703355256e-05, "epoch": 0.0761904761904762, "percentage": 2.56, "elapsed_time": "0:01:41", "remaining_time": "1:04:29", "throughput": 2074.94, "total_tokens": 211280}
+{"current_steps": 10, "total_steps": 195, "loss": 0.2409, "lr": 4.967625656594782e-05, "epoch": 0.1523809523809524, "percentage": 5.13, "elapsed_time": "0:03:23", "remaining_time": "1:02:45", "throughput": 2059.35, "total_tokens": 419136}
+{"current_steps": 15, "total_steps": 195, "loss": 0.1711, "lr": 4.92735454356513e-05, "epoch": 0.22857142857142856, "percentage": 7.69, "elapsed_time": "0:05:09", "remaining_time": "1:01:57", "throughput": 2038.91, "total_tokens": 631632}
+{"current_steps": 20, "total_steps": 195, "loss": 0.1228, "lr": 4.8713411048678635e-05, "epoch": 0.3047619047619048, "percentage": 10.26, "elapsed_time": "0:06:51", "remaining_time": "1:00:00", "throughput": 2033.99, "total_tokens": 836896}
+{"current_steps": 25, "total_steps": 195, "loss": 0.0945, "lr": 4.799948609147061e-05, "epoch": 0.38095238095238093, "percentage": 12.82, "elapsed_time": "0:08:33", "remaining_time": "0:58:10", "throughput": 2030.92, "total_tokens": 1042416}
+{"current_steps": 30, "total_steps": 195, "loss": 0.0873, "lr": 4.713640064133025e-05, "epoch": 0.45714285714285713, "percentage": 15.38, "elapsed_time": "0:10:13", "remaining_time": "0:56:14", "throughput": 2028.08, "total_tokens": 1244384}
+{"current_steps": 35, "total_steps": 195, "loss": 0.0635, "lr": 4.6129752138594874e-05, "epoch": 0.5333333333333333, "percentage": 17.95, "elapsed_time": "0:11:51", "remaining_time": "0:54:12", "throughput": 2030.96, "total_tokens": 1445168}
+{"current_steps": 40, "total_steps": 195, "loss": 0.0784, "lr": 4.498606908508754e-05, "epoch": 0.6095238095238096, "percentage": 20.51, "elapsed_time": "0:13:33", "remaining_time": "0:52:34", "throughput": 2029.27, "total_tokens": 1651760}
+{"current_steps": 45, "total_steps": 195, "loss": 0.0634, "lr": 4.371276870427753e-05, "epoch": 0.6857142857142857, "percentage": 23.08, "elapsed_time": "0:15:10", "remaining_time": "0:50:33", "throughput": 2032.57, "total_tokens": 1849920}
+{"current_steps": 50, "total_steps": 195, "loss": 0.0715, "lr": 4.231810883773999e-05, "epoch": 0.7619047619047619, "percentage": 25.64, "elapsed_time": "0:16:52", "remaining_time": "0:48:57", "throughput": 2030.24, "total_tokens": 2056592}
+{"current_steps": 55, "total_steps": 195, "loss": 0.0588, "lr": 4.0811134389884433e-05, "epoch": 0.8380952380952381, "percentage": 28.21, "elapsed_time": "0:18:35", "remaining_time": "0:47:19", "throughput": 2030.66, "total_tokens": 2265056}
+{"current_steps": 60, "total_steps": 195, "loss": 0.0634, "lr": 3.920161866827889e-05, "epoch": 0.9142857142857143, "percentage": 30.77, "elapsed_time": "0:20:17", "remaining_time": "0:45:38", "throughput": 2030.21, "total_tokens": 2471328}
+{"current_steps": 65, "total_steps": 195, "loss": 0.0582, "lr": 3.7500000000000003e-05, "epoch": 0.9904761904761905, "percentage": 33.33, "elapsed_time": "0:21:57", "remaining_time": "0:43:55", "throughput": 2030.85, "total_tokens": 2676640}
+{"current_steps": 70, "total_steps": 195, "loss": 0.0687, "lr": 3.5717314035076355e-05, "epoch": 1.0761904761904761, "percentage": 35.9, "elapsed_time": "0:23:47", "remaining_time": "0:42:29", "throughput": 2029.43, "total_tokens": 2897712}
+{"current_steps": 75, "total_steps": 195, "loss": 0.0539, "lr": 3.386512217606339e-05, "epoch": 1.1523809523809523, "percentage": 38.46, "elapsed_time": "0:25:19", "remaining_time": "0:40:30", "throughput": 2033.85, "total_tokens": 3089808}
+{"current_steps": 80, "total_steps": 195, "loss": 0.0545, "lr": 3.195543659791132e-05, "epoch": 1.2285714285714286, "percentage": 41.03, "elapsed_time": "0:27:03", "remaining_time": "0:38:53", "throughput": 2030.86, "total_tokens": 3296576}
+{"current_steps": 85, "total_steps": 195, "loss": 0.0446, "lr": 3.0000642344401113e-05, "epoch": 1.3047619047619048, "percentage": 43.59, "elapsed_time": "0:28:48", "remaining_time": "0:37:16", "throughput": 2030.02, "total_tokens": 3508416}
+{"current_steps": 90, "total_steps": 195, "loss": 0.0535, "lr": 2.8013417006383076e-05, "epoch": 1.380952380952381, "percentage": 46.15, "elapsed_time": "0:30:25", "remaining_time": "0:35:29", "throughput": 2030.47, "total_tokens": 3706512}
+{"current_steps": 95, "total_steps": 195, "loss": 0.0488, "lr": 2.600664850273538e-05, "epoch": 1.457142857142857, "percentage": 48.72, "elapsed_time": "0:32:09", "remaining_time": "0:33:50", "throughput": 2028.22, "total_tokens": 3913120}
+{"current_steps": 100, "total_steps": 195, "loss": 0.044, "lr": 2.399335149726463e-05, "epoch": 1.5333333333333332, "percentage": 51.28, "elapsed_time": "0:33:54", "remaining_time": "0:32:12", "throughput": 2026.11, "total_tokens": 4121248}
+{"current_steps": 105, "total_steps": 195, "loss": 0.0479, "lr": 2.1986582993616926e-05, "epoch": 1.6095238095238096, "percentage": 53.85, "elapsed_time": "0:35:37", "remaining_time": "0:30:32", "throughput": 2024.85, "total_tokens": 4327824}
+{"current_steps": 110, "total_steps": 195, "loss": 0.0441, "lr": 1.9999357655598893e-05, "epoch": 1.6857142857142857, "percentage": 56.41, "elapsed_time": "0:37:22", "remaining_time": "0:28:52", "throughput": 2023.58, "total_tokens": 4537936}
+{"current_steps": 115, "total_steps": 195, "loss": 0.0428, "lr": 1.8044563402088684e-05, "epoch": 1.7619047619047619, "percentage": 58.97, "elapsed_time": "0:39:04", "remaining_time": "0:27:11", "throughput": 2023.28, "total_tokens": 4744432}
+{"current_steps": 120, "total_steps": 195, "loss": 0.0425, "lr": 1.613487782393661e-05, "epoch": 1.8380952380952382, "percentage": 61.54, "elapsed_time": "0:40:48", "remaining_time": "0:25:30", "throughput": 2023.49, "total_tokens": 4955296}
+{"current_steps": 125, "total_steps": 195, "loss": 0.0437, "lr": 1.4282685964923642e-05, "epoch": 1.9142857142857141, "percentage": 64.1, "elapsed_time": "0:42:24", "remaining_time": "0:23:45", "throughput": 2025.72, "total_tokens": 5154784}
+{"current_steps": 130, "total_steps": 195, "loss": 0.039, "lr": 1.2500000000000006e-05, "epoch": 1.9904761904761905, "percentage": 66.67, "elapsed_time": "0:44:03", "remaining_time": "0:22:01", "throughput": 2026.78, "total_tokens": 5358160}
+{"current_steps": 135, "total_steps": 195, "loss": 0.0465, "lr": 1.0798381331721109e-05, "epoch": 2.0761904761904764, "percentage": 69.23, "elapsed_time": "0:45:52", "remaining_time": "0:20:23", "throughput": 2026.32, "total_tokens": 5577808}
+{"current_steps": 140, "total_steps": 195, "loss": 0.0451, "lr": 9.18886561011557e-06, "epoch": 2.1523809523809523, "percentage": 71.79, "elapsed_time": "0:47:27", "remaining_time": "0:18:38", "throughput": 2027.94, "total_tokens": 5775552}
+{"current_steps": 145, "total_steps": 195, "loss": 0.0385, "lr": 7.681891162260015e-06, "epoch": 2.2285714285714286, "percentage": 74.36, "elapsed_time": "0:49:14", "remaining_time": "0:16:58", "throughput": 2026.17, "total_tokens": 5986112}
+{"current_steps": 150, "total_steps": 195, "loss": 0.0506, "lr": 6.28723129572247e-06, "epoch": 2.3047619047619046, "percentage": 76.92, "elapsed_time": "0:50:51", "remaining_time": "0:15:15", "throughput": 2027.23, "total_tokens": 6185568}
+{"current_steps": 155, "total_steps": 195, "loss": 0.0471, "lr": 5.013930914912476e-06, "epoch": 2.380952380952381, "percentage": 79.49, "elapsed_time": "0:52:30", "remaining_time": "0:13:33", "throughput": 2027.9, "total_tokens": 6389392}
+{"current_steps": 160, "total_steps": 195, "loss": 0.0383, "lr": 3.8702478614051355e-06, "epoch": 2.4571428571428573, "percentage": 82.05, "elapsed_time": "0:54:11", "remaining_time": "0:11:51", "throughput": 2028.25, "total_tokens": 6594256}
+{"current_steps": 165, "total_steps": 195, "loss": 0.036, "lr": 2.8635993586697553e-06, "epoch": 2.533333333333333, "percentage": 84.62, "elapsed_time": "0:55:53", "remaining_time": "0:10:09", "throughput": 2028.63, "total_tokens": 6802240}
+{"current_steps": 170, "total_steps": 195, "loss": 0.0381, "lr": 2.0005139085293945e-06, "epoch": 2.6095238095238096, "percentage": 87.18, "elapsed_time": "0:57:32", "remaining_time": "0:08:27", "throughput": 2028.93, "total_tokens": 7004816}
+{"current_steps": 175, "total_steps": 195, "loss": 0.0415, "lr": 1.286588951321363e-06, "epoch": 2.685714285714286, "percentage": 89.74, "elapsed_time": "0:59:12", "remaining_time": "0:06:46", "throughput": 2028.93, "total_tokens": 7208608}
+{"current_steps": 180, "total_steps": 195, "loss": 0.0397, "lr": 7.264545643486997e-07, "epoch": 2.761904761904762, "percentage": 92.31, "elapsed_time": "1:00:58", "remaining_time": "0:05:04", "throughput": 2027.85, "total_tokens": 7418320}
+{"current_steps": 185, "total_steps": 195, "loss": 0.0421, "lr": 3.237434340521789e-07, "epoch": 2.8380952380952382, "percentage": 94.87, "elapsed_time": "1:02:48", "remaining_time": "0:03:23", "throughput": 2025.27, "total_tokens": 7633248}
+{"current_steps": 190, "total_steps": 195, "loss": 0.0378, "lr": 8.106729664475176e-08, "epoch": 2.914285714285714, "percentage": 97.44, "elapsed_time": "1:04:29", "remaining_time": "0:01:41", "throughput": 2025.73, "total_tokens": 7838576}
+{"current_steps": 195, "total_steps": 195, "loss": 0.047, "lr": 0.0, "epoch": 2.9904761904761905, "percentage": 100.0, "elapsed_time": "1:06:07", "remaining_time": "0:00:00", "throughput": 2026.33, "total_tokens": 8039616}
+{"current_steps": 195, "total_steps": 195, "epoch": 2.9904761904761905, "percentage": 100.0, "elapsed_time": "1:06:10", "remaining_time": "0:00:00", "throughput": 2025.01, "total_tokens": 8039616}

trainer_state.json

@@ -0,0 +1,355 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9904761904761905,
+  "eval_steps": 500,
+  "global_step": 195,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0761904761904762,
+      "grad_norm": 0.2635799050331116,
+      "learning_rate": 4.9918932703355256e-05,
+      "loss": 0.3154,
+      "num_input_tokens_seen": 211280,
+      "step": 5
+    },
+    {
+      "epoch": 0.1523809523809524,
+      "grad_norm": 0.2279728353023529,
+      "learning_rate": 4.967625656594782e-05,
+      "loss": 0.2409,
+      "num_input_tokens_seen": 419136,
+      "step": 10
+    },
+    {
+      "epoch": 0.22857142857142856,
+      "grad_norm": 0.22582456469535828,
+      "learning_rate": 4.92735454356513e-05,
+      "loss": 0.1711,
+      "num_input_tokens_seen": 631632,
+      "step": 15
+    },
+    {
+      "epoch": 0.3047619047619048,
+      "grad_norm": 0.14574189484119415,
+      "learning_rate": 4.8713411048678635e-05,
+      "loss": 0.1228,
+      "num_input_tokens_seen": 836896,
+      "step": 20
+    },
+    {
+      "epoch": 0.38095238095238093,
+      "grad_norm": 0.12243571132421494,
+      "learning_rate": 4.799948609147061e-05,
+      "loss": 0.0945,
+      "num_input_tokens_seen": 1042416,
+      "step": 25
+    },
+    {
+      "epoch": 0.45714285714285713,
+      "grad_norm": 0.12587733566761017,
+      "learning_rate": 4.713640064133025e-05,
+      "loss": 0.0873,
+      "num_input_tokens_seen": 1244384,
+      "step": 30
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.09762891381978989,
+      "learning_rate": 4.6129752138594874e-05,
+      "loss": 0.0635,
+      "num_input_tokens_seen": 1445168,
+      "step": 35
+    },
+    {
+      "epoch": 0.6095238095238096,
+      "grad_norm": 0.10953030735254288,
+      "learning_rate": 4.498606908508754e-05,
+      "loss": 0.0784,
+      "num_input_tokens_seen": 1651760,
+      "step": 40
+    },
+    {
+      "epoch": 0.6857142857142857,
+      "grad_norm": 0.12629230320453644,
+      "learning_rate": 4.371276870427753e-05,
+      "loss": 0.0634,
+      "num_input_tokens_seen": 1849920,
+      "step": 45
+    },
+    {
+      "epoch": 0.7619047619047619,
+      "grad_norm": 0.12405364215373993,
+      "learning_rate": 4.231810883773999e-05,
+      "loss": 0.0715,
+      "num_input_tokens_seen": 2056592,
+      "step": 50
+    },
+    {
+      "epoch": 0.8380952380952381,
+      "grad_norm": 0.07898106426000595,
+      "learning_rate": 4.0811134389884433e-05,
+      "loss": 0.0588,
+      "num_input_tokens_seen": 2265056,
+      "step": 55
+    },
+    {
+      "epoch": 0.9142857142857143,
+      "grad_norm": 0.10841359943151474,
+      "learning_rate": 3.920161866827889e-05,
+      "loss": 0.0634,
+      "num_input_tokens_seen": 2471328,
+      "step": 60
+    },
+    {
+      "epoch": 0.9904761904761905,
+      "grad_norm": 0.1293160319328308,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.0582,
+      "num_input_tokens_seen": 2676640,
+      "step": 65
+    },
+    {
+      "epoch": 1.0761904761904761,
+      "grad_norm": 0.11754991114139557,
+      "learning_rate": 3.5717314035076355e-05,
+      "loss": 0.0687,
+      "num_input_tokens_seen": 2897712,
+      "step": 70
+    },
+    {
+      "epoch": 1.1523809523809523,
+      "grad_norm": 0.10079122334718704,
+      "learning_rate": 3.386512217606339e-05,
+      "loss": 0.0539,
+      "num_input_tokens_seen": 3089808,
+      "step": 75
+    },
+    {
+      "epoch": 1.2285714285714286,
+      "grad_norm": 0.10278293490409851,
+      "learning_rate": 3.195543659791132e-05,
+      "loss": 0.0545,
+      "num_input_tokens_seen": 3296576,
+      "step": 80
+    },
+    {
+      "epoch": 1.3047619047619048,
+      "grad_norm": 0.10142832249403,
+      "learning_rate": 3.0000642344401113e-05,
+      "loss": 0.0446,
+      "num_input_tokens_seen": 3508416,
+      "step": 85
+    },
+    {
+      "epoch": 1.380952380952381,
+      "grad_norm": 0.21511930227279663,
+      "learning_rate": 2.8013417006383076e-05,
+      "loss": 0.0535,
+      "num_input_tokens_seen": 3706512,
+      "step": 90
+    },
+    {
+      "epoch": 1.457142857142857,
+      "grad_norm": 0.10380080342292786,
+      "learning_rate": 2.600664850273538e-05,
+      "loss": 0.0488,
+      "num_input_tokens_seen": 3913120,
+      "step": 95
+    },
+    {
+      "epoch": 1.5333333333333332,
+      "grad_norm": 0.11608559638261795,
+      "learning_rate": 2.399335149726463e-05,
+      "loss": 0.044,
+      "num_input_tokens_seen": 4121248,
+      "step": 100
+    },
+    {
+      "epoch": 1.6095238095238096,
+      "grad_norm": 0.11765897274017334,
+      "learning_rate": 2.1986582993616926e-05,
+      "loss": 0.0479,
+      "num_input_tokens_seen": 4327824,
+      "step": 105
+    },
+    {
+      "epoch": 1.6857142857142857,
+      "grad_norm": 0.0989590659737587,
+      "learning_rate": 1.9999357655598893e-05,
+      "loss": 0.0441,
+      "num_input_tokens_seen": 4537936,
+      "step": 110
+    },
+    {
+      "epoch": 1.7619047619047619,
+      "grad_norm": 0.1432141810655594,
+      "learning_rate": 1.8044563402088684e-05,
+      "loss": 0.0428,
+      "num_input_tokens_seen": 4744432,
+      "step": 115
+    },
+    {
+      "epoch": 1.8380952380952382,
+      "grad_norm": 0.08667314797639847,
+      "learning_rate": 1.613487782393661e-05,
+      "loss": 0.0425,
+      "num_input_tokens_seen": 4955296,
+      "step": 120
+    },
+    {
+      "epoch": 1.9142857142857141,
+      "grad_norm": 0.12118078023195267,
+      "learning_rate": 1.4282685964923642e-05,
+      "loss": 0.0437,
+      "num_input_tokens_seen": 5154784,
+      "step": 125
+    },
+    {
+      "epoch": 1.9904761904761905,
+      "grad_norm": 0.09234391897916794,
+      "learning_rate": 1.2500000000000006e-05,
+      "loss": 0.039,
+      "num_input_tokens_seen": 5358160,
+      "step": 130
+    },
+    {
+      "epoch": 2.0761904761904764,
+      "grad_norm": 0.08275337517261505,
+      "learning_rate": 1.0798381331721109e-05,
+      "loss": 0.0465,
+      "num_input_tokens_seen": 5577808,
+      "step": 135
+    },
+    {
+      "epoch": 2.1523809523809523,
+      "grad_norm": 0.08790527284145355,
+      "learning_rate": 9.18886561011557e-06,
+      "loss": 0.0451,
+      "num_input_tokens_seen": 5775552,
+      "step": 140
+    },
+    {
+      "epoch": 2.2285714285714286,
+      "grad_norm": 0.0865679457783699,
+      "learning_rate": 7.681891162260015e-06,
+      "loss": 0.0385,
+      "num_input_tokens_seen": 5986112,
+      "step": 145
+    },
+    {
+      "epoch": 2.3047619047619046,
+      "grad_norm": 0.09322898089885712,
+      "learning_rate": 6.28723129572247e-06,
+      "loss": 0.0506,
+      "num_input_tokens_seen": 6185568,
+      "step": 150
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.13761946558952332,
+      "learning_rate": 5.013930914912476e-06,
+      "loss": 0.0471,
+      "num_input_tokens_seen": 6389392,
+      "step": 155
+    },
+    {
+      "epoch": 2.4571428571428573,
+      "grad_norm": 0.1628292351961136,
+      "learning_rate": 3.8702478614051355e-06,
+      "loss": 0.0383,
+      "num_input_tokens_seen": 6594256,
+      "step": 160
+    },
+    {
+      "epoch": 2.533333333333333,
+      "grad_norm": 0.09071007370948792,
+      "learning_rate": 2.8635993586697553e-06,
+      "loss": 0.036,
+      "num_input_tokens_seen": 6802240,
+      "step": 165
+    },
+    {
+      "epoch": 2.6095238095238096,
+      "grad_norm": 0.09868919104337692,
+      "learning_rate": 2.0005139085293945e-06,
+      "loss": 0.0381,
+      "num_input_tokens_seen": 7004816,
+      "step": 170
+    },
+    {
+      "epoch": 2.685714285714286,
+      "grad_norm": 0.11255531758069992,
+      "learning_rate": 1.286588951321363e-06,
+      "loss": 0.0415,
+      "num_input_tokens_seen": 7208608,
+      "step": 175
+    },
+    {
+      "epoch": 2.761904761904762,
+      "grad_norm": 0.09013009071350098,
+      "learning_rate": 7.264545643486997e-07,
+      "loss": 0.0397,
+      "num_input_tokens_seen": 7418320,
+      "step": 180
+    },
+    {
+      "epoch": 2.8380952380952382,
+      "grad_norm": 0.10500302165746689,
+      "learning_rate": 3.237434340521789e-07,
+      "loss": 0.0421,
+      "num_input_tokens_seen": 7633248,
+      "step": 185
+    },
+    {
+      "epoch": 2.914285714285714,
+      "grad_norm": 0.09550434350967407,
+      "learning_rate": 8.106729664475176e-08,
+      "loss": 0.0378,
+      "num_input_tokens_seen": 7838576,
+      "step": 190
+    },
+    {
+      "epoch": 2.9904761904761905,
+      "grad_norm": 0.10578631609678268,
+      "learning_rate": 0.0,
+      "loss": 0.047,
+      "num_input_tokens_seen": 8039616,
+      "step": 195
+    },
+    {
+      "epoch": 2.9904761904761905,
+      "num_input_tokens_seen": 8039616,
+      "step": 195,
+      "total_flos": 3.746792787278561e+17,
+      "train_loss": 0.06834620604148278,
+      "train_runtime": 3970.1665,
+      "train_samples_per_second": 0.793,
+      "train_steps_per_second": 0.049
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 195,
+  "num_input_tokens_seen": 8039616,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.746792787278561e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64b42ad3b4b8b79db7da6f9ec535c5eca035fc96c8a7dc13db39845793113a1c
+size 5688

training_args.yaml

@@ -0,0 +1,33 @@
+bf16: true
+cutoff_len: 34114
+dataset: BsKIE3
+dataset_dir: data
+ddp_timeout: 180000000
+do_train: true
+finetuning_type: lora
+flash_attn: auto
+gradient_accumulation_steps: 8
+include_num_input_tokens_seen: true
+learning_rate: 5.0e-05
+logging_steps: 5
+lora_alpha: 16
+lora_dropout: 0
+lora_rank: 8
+lora_target: all
+lr_scheduler_type: cosine
+max_grad_norm: 1.0
+max_samples: 100000
+model_name_or_path: Qwen/Qwen2.5-VL-7B-Instruct
+num_train_epochs: 3.0
+optim: adamw_torch
+output_dir: saves/Qwen2.5-VL-7B-Instruct/lora/BS_riche_lora
+packing: false
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+report_to: none
+save_steps: 100
+stage: sft
+template: qwen2_vl
+trust_remote_code: true
+warmup_steps: 0