Init model.

README.md

@@ -0,0 +1,61 @@
+---
+library_name: peft
+license: other
+base_model: llava-hf/llava-1.5-7b-hf
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: train_zh_1_fold_4_epochs
+  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. -->
+
+# train_zh_1_fold_4_epochs
+
+This model is a fine-tuned version of [llava-hf/llava-1.5-7b-hf](https://huggingface.co/llava-hf/llava-1.5-7b-hf) on the sticker_labels_kfolds_train_zh_1 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
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 32
+- total_eval_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: 4.0
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.15.2
+- Transformers 4.52.1
+- Pytorch 2.7.0+cu126
+- Datasets 3.6.0
+- Tokenizers 0.21.1

adapter_config.json

@@ -0,0 +1,132 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "llava-hf/llava-1.5-7b-hf",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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_bias": false,
+  "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": [
+    "language_model.layers.15.self_attn.v_proj",
+    "language_model.layers.15.self_attn.k_proj",
+    "language_model.layers.2.self_attn.v_proj",
+    "language_model.layers.11.self_attn.v_proj",
+    "language_model.layers.2.self_attn.q_proj",
+    "language_model.layers.23.self_attn.k_proj",
+    "28.self_attn.q_proj",
+    "language_model.layers.9.self_attn.v_proj",
+    "31.self_attn.k_proj",
+    "o_proj",
+    "language_model.layers.13.self_attn.k_proj",
+    "language_model.layers.5.self_attn.v_proj",
+    "language_model.layers.3.self_attn.k_proj",
+    "language_model.layers.21.self_attn.k_proj",
+    "language_model.layers.14.self_attn.v_proj",
+    "language_model.layers.6.self_attn.k_proj",
+    "language_model.layers.4.self_attn.q_proj",
+    "24.self_attn.v_proj",
+    "27.self_attn.q_proj",
+    "language_model.layers.10.self_attn.v_proj",
+    "language_model.layers.11.self_attn.k_proj",
+    "language_model.layers.12.self_attn.q_proj",
+    "27.self_attn.k_proj",
+    "gate_proj",
+    "language_model.layers.12.self_attn.v_proj",
+    "language_model.layers.9.self_attn.q_proj",
+    "language_model.layers.18.self_attn.v_proj",
+    "24.self_attn.q_proj",
+    "language_model.layers.7.self_attn.k_proj",
+    "language_model.layers.14.self_attn.q_proj",
+    "language_model.layers.22.self_attn.k_proj",
+    "31.self_attn.q_proj",
+    "language_model.layers.13.self_attn.v_proj",
+    "language_model.layers.16.self_attn.v_proj",
+    "language_model.layers.3.self_attn.q_proj",
+    "language_model.layers.19.self_attn.q_proj",
+    "29.self_attn.q_proj",
+    "27.self_attn.v_proj",
+    "language_model.layers.19.self_attn.v_proj",
+    "26.self_attn.q_proj",
+    "language_model.layers.10.self_attn.k_proj",
+    "language_model.layers.0.self_attn.v_proj",
+    "30.self_attn.v_proj",
+    "language_model.layers.18.self_attn.k_proj",
+    "language_model.layers.7.self_attn.v_proj",
+    "language_model.layers.8.self_attn.q_proj",
+    "language_model.layers.20.self_attn.q_proj",
+    "language_model.layers.11.self_attn.q_proj",
+    "28.self_attn.k_proj",
+    "language_model.layers.23.self_attn.v_proj",
+    "language_model.layers.20.self_attn.v_proj",
+    "language_model.layers.0.self_attn.q_proj",
+    "language_model.layers.7.self_attn.q_proj",
+    "language_model.layers.0.self_attn.k_proj",
+    "25.self_attn.k_proj",
+    "language_model.layers.12.self_attn.k_proj",
+    "29.self_attn.k_proj",
+    "language_model.layers.3.self_attn.v_proj",
+    "language_model.layers.18.self_attn.q_proj",
+    "language_model.layers.22.self_attn.v_proj",
+    "26.self_attn.k_proj",
+    "language_model.layers.4.self_attn.k_proj",
+    "language_model.layers.22.self_attn.q_proj",
+    "30.self_attn.q_proj",
+    "language_model.layers.23.self_attn.q_proj",
+    "language_model.layers.1.self_attn.q_proj",
+    "language_model.layers.16.self_attn.k_proj",
+    "language_model.layers.13.self_attn.q_proj",
+    "down_proj",
+    "language_model.layers.10.self_attn.q_proj",
+    "language_model.layers.1.self_attn.k_proj",
+    "language_model.layers.2.self_attn.k_proj",
+    "31.self_attn.v_proj",
+    "language_model.layers.21.self_attn.q_proj",
+    "28.self_attn.v_proj",
+    "language_model.layers.15.self_attn.q_proj",
+    "language_model.layers.6.self_attn.q_proj",
+    "25.self_attn.v_proj",
+    "26.self_attn.v_proj",
+    "language_model.layers.20.self_attn.k_proj",
+    "language_model.layers.6.self_attn.v_proj",
+    "language_model.layers.17.self_attn.q_proj",
+    "language_model.layers.9.self_attn.k_proj",
+    "language_model.layers.5.self_attn.k_proj",
+    "language_model.layers.8.self_attn.k_proj",
+    "language_model.layers.17.self_attn.v_proj",
+    "language_model.layers.21.self_attn.v_proj",
+    "language_model.layers.16.self_attn.q_proj",
+    "language_model.layers.14.self_attn.k_proj",
+    "language_model.layers.5.self_attn.q_proj",
+    "29.self_attn.v_proj",
+    "up_proj",
+    "24.self_attn.k_proj",
+    "30.self_attn.k_proj",
+    "language_model.layers.8.self_attn.v_proj",
+    "language_model.layers.4.self_attn.v_proj",
+    "language_model.layers.17.self_attn.k_proj",
+    "25.self_attn.q_proj",
+    "language_model.layers.1.self_attn.v_proj",
+    "language_model.layers.19.self_attn.k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1353f770eadba537a1aeef0aed0ad61ec839df901a204edf59c09b8f0dae88e3
+size 80019840

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "<image>": 32000,
+  "<pad>": 32001
+}

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.0,
+    "num_input_tokens_seen": 1471136,
+    "total_flos": 6.135952763047117e+16,
+    "train_loss": 1.7421463951468468,
+    "train_runtime": 339.8451,
+    "train_samples_per_second": 5.756,
+    "train_steps_per_second": 0.188
+}

chat_template.jinja

@@ -0,0 +1 @@
+{% set system_message = 'A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user\'s questions.' %}{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'USER: ' + content + ' ASSISTANT:' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' }}{% endif %}{% endfor %}

checkpoint-64/README.md

@@ -0,0 +1,202 @@
+---
+base_model: llava-hf/llava-1.5-7b-hf
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2

checkpoint-64/adapter_config.json

@@ -0,0 +1,132 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "llava-hf/llava-1.5-7b-hf",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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_bias": false,
+  "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": [
+    "language_model.layers.15.self_attn.v_proj",
+    "language_model.layers.15.self_attn.k_proj",
+    "language_model.layers.2.self_attn.v_proj",
+    "language_model.layers.11.self_attn.v_proj",
+    "language_model.layers.2.self_attn.q_proj",
+    "language_model.layers.23.self_attn.k_proj",
+    "28.self_attn.q_proj",
+    "language_model.layers.9.self_attn.v_proj",
+    "31.self_attn.k_proj",
+    "o_proj",
+    "language_model.layers.13.self_attn.k_proj",
+    "language_model.layers.5.self_attn.v_proj",
+    "language_model.layers.3.self_attn.k_proj",
+    "language_model.layers.21.self_attn.k_proj",
+    "language_model.layers.14.self_attn.v_proj",
+    "language_model.layers.6.self_attn.k_proj",
+    "language_model.layers.4.self_attn.q_proj",
+    "24.self_attn.v_proj",
+    "27.self_attn.q_proj",
+    "language_model.layers.10.self_attn.v_proj",
+    "language_model.layers.11.self_attn.k_proj",
+    "language_model.layers.12.self_attn.q_proj",
+    "27.self_attn.k_proj",
+    "gate_proj",
+    "language_model.layers.12.self_attn.v_proj",
+    "language_model.layers.9.self_attn.q_proj",
+    "language_model.layers.18.self_attn.v_proj",
+    "24.self_attn.q_proj",
+    "language_model.layers.7.self_attn.k_proj",
+    "language_model.layers.14.self_attn.q_proj",
+    "language_model.layers.22.self_attn.k_proj",
+    "31.self_attn.q_proj",
+    "language_model.layers.13.self_attn.v_proj",
+    "language_model.layers.16.self_attn.v_proj",
+    "language_model.layers.3.self_attn.q_proj",
+    "language_model.layers.19.self_attn.q_proj",
+    "29.self_attn.q_proj",
+    "27.self_attn.v_proj",
+    "language_model.layers.19.self_attn.v_proj",
+    "26.self_attn.q_proj",
+    "language_model.layers.10.self_attn.k_proj",
+    "language_model.layers.0.self_attn.v_proj",
+    "30.self_attn.v_proj",
+    "language_model.layers.18.self_attn.k_proj",
+    "language_model.layers.7.self_attn.v_proj",
+    "language_model.layers.8.self_attn.q_proj",
+    "language_model.layers.20.self_attn.q_proj",
+    "language_model.layers.11.self_attn.q_proj",
+    "28.self_attn.k_proj",
+    "language_model.layers.23.self_attn.v_proj",
+    "language_model.layers.20.self_attn.v_proj",
+    "language_model.layers.0.self_attn.q_proj",
+    "language_model.layers.7.self_attn.q_proj",
+    "language_model.layers.0.self_attn.k_proj",
+    "25.self_attn.k_proj",
+    "language_model.layers.12.self_attn.k_proj",
+    "29.self_attn.k_proj",
+    "language_model.layers.3.self_attn.v_proj",
+    "language_model.layers.18.self_attn.q_proj",
+    "language_model.layers.22.self_attn.v_proj",
+    "26.self_attn.k_proj",
+    "language_model.layers.4.self_attn.k_proj",
+    "language_model.layers.22.self_attn.q_proj",
+    "30.self_attn.q_proj",
+    "language_model.layers.23.self_attn.q_proj",
+    "language_model.layers.1.self_attn.q_proj",
+    "language_model.layers.16.self_attn.k_proj",
+    "language_model.layers.13.self_attn.q_proj",
+    "down_proj",
+    "language_model.layers.10.self_attn.q_proj",
+    "language_model.layers.1.self_attn.k_proj",
+    "language_model.layers.2.self_attn.k_proj",
+    "31.self_attn.v_proj",
+    "language_model.layers.21.self_attn.q_proj",
+    "28.self_attn.v_proj",
+    "language_model.layers.15.self_attn.q_proj",
+    "language_model.layers.6.self_attn.q_proj",
+    "25.self_attn.v_proj",
+    "26.self_attn.v_proj",
+    "language_model.layers.20.self_attn.k_proj",
+    "language_model.layers.6.self_attn.v_proj",
+    "language_model.layers.17.self_attn.q_proj",
+    "language_model.layers.9.self_attn.k_proj",
+    "language_model.layers.5.self_attn.k_proj",
+    "language_model.layers.8.self_attn.k_proj",
+    "language_model.layers.17.self_attn.v_proj",
+    "language_model.layers.21.self_attn.v_proj",
+    "language_model.layers.16.self_attn.q_proj",
+    "language_model.layers.14.self_attn.k_proj",
+    "language_model.layers.5.self_attn.q_proj",
+    "29.self_attn.v_proj",
+    "up_proj",
+    "24.self_attn.k_proj",
+    "30.self_attn.k_proj",
+    "language_model.layers.8.self_attn.v_proj",
+    "language_model.layers.4.self_attn.v_proj",
+    "language_model.layers.17.self_attn.k_proj",
+    "25.self_attn.q_proj",
+    "language_model.layers.1.self_attn.v_proj",
+    "language_model.layers.19.self_attn.k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-64/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1353f770eadba537a1aeef0aed0ad61ec839df901a204edf59c09b8f0dae88e3
+size 80019840

checkpoint-64/added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "<image>": 32000,
+  "<pad>": 32001
+}

checkpoint-64/chat_template.jinja

@@ -0,0 +1,2 @@
+{% for message in messages %}{% if message['role'] != 'system' %}{{ message['role'].upper() + ': '}}{% endif %}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>
+' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ content['text'] + ' '}}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ content['text'] + ' '}}{% endgeneration %}{% endfor %}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'ASSISTANT:' }}{% endif %}

checkpoint-64/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1828e38a4c554859aa0328b980b1470cea518a17fe1b5fbedf8861363884385
+size 160285219

checkpoint-64/preprocessor_config.json

@@ -0,0 +1,28 @@
+{
+  "crop_size": {
+    "height": 336,
+    "width": 336
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "processor_class": "LlavaProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 336
+  }
+}

checkpoint-64/processor_config.json

@@ -0,0 +1,7 @@
+{
+  "image_token": "<image>",
+  "num_additional_image_tokens": 1,
+  "patch_size": 14,
+  "processor_class": "LlavaProcessor",
+  "vision_feature_select_strategy": "default"
+}

checkpoint-64/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02ea5dcfd1b4a49b41b4fa01a8b24bba6186957162c3fd555ebff28620c7268b
+size 14917

checkpoint-64/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2a5af18bb5eae8b7fd6bdef66259014d98ba87ffb16d614bba38f2c32030798
+size 14917

checkpoint-64/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcbcddf99e9968b159b35666b0f12cc660b256795ba1422c789366592404af5e
+size 1465

checkpoint-64/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-64/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

checkpoint-64/tokenizer_config.json

@@ -0,0 +1,65 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32000": {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {
+    "image_token": "<image>"
+  },
+  "image_token": "<image>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "padding_side": "right",
+  "processor_class": "LlavaProcessor",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "trust_remote_code": false,
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-64/trainer_state.json

@@ -0,0 +1,154 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 64,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3252032520325203,
+      "grad_norm": 1.3156858682632446,
+      "learning_rate": 4.951963201008076e-05,
+      "loss": 2.2304,
+      "num_input_tokens_seen": 119024,
+      "step": 5,
+      "train_runtime": 28.3475,
+      "train_tokens_per_second": 4198.75
+    },
+    {
+      "epoch": 0.6504065040650406,
+      "grad_norm": 0.7530735731124878,
+      "learning_rate": 4.759973232808609e-05,
+      "loss": 2.147,
+      "num_input_tokens_seen": 238880,
+      "step": 10,
+      "train_runtime": 55.416,
+      "train_tokens_per_second": 4310.67
+    },
+    {
+      "epoch": 0.975609756097561,
+      "grad_norm": 0.6344960927963257,
+      "learning_rate": 4.4325261334068426e-05,
+      "loss": 1.9562,
+      "num_input_tokens_seen": 358608,
+      "step": 15,
+      "train_runtime": 82.5363,
+      "train_tokens_per_second": 4344.849
+    },
+    {
+      "epoch": 1.2601626016260163,
+      "grad_norm": 0.5793188214302063,
+      "learning_rate": 3.9892482612310836e-05,
+      "loss": 1.5887,
+      "num_input_tokens_seen": 463568,
+      "step": 20,
+      "train_runtime": 106.4581,
+      "train_tokens_per_second": 4354.466
+    },
+    {
+      "epoch": 1.5853658536585367,
+      "grad_norm": 0.5236935615539551,
+      "learning_rate": 3.456708580912725e-05,
+      "loss": 1.7952,
+      "num_input_tokens_seen": 583360,
+      "step": 25,
+      "train_runtime": 133.8831,
+      "train_tokens_per_second": 4357.233
+    },
+    {
+      "epoch": 1.910569105691057,
+      "grad_norm": 0.6524015665054321,
+      "learning_rate": 2.8668261861384045e-05,
+      "loss": 1.7717,
+      "num_input_tokens_seen": 702656,
+      "step": 30,
+      "train_runtime": 161.3217,
+      "train_tokens_per_second": 4355.62
+    },
+    {
+      "epoch": 2.1951219512195124,
+      "grad_norm": 0.5044527053833008,
+      "learning_rate": 2.2549571491760986e-05,
+      "loss": 1.5037,
+      "num_input_tokens_seen": 807312,
+      "step": 35,
+      "train_runtime": 185.3057,
+      "train_tokens_per_second": 4356.649
+    },
+    {
+      "epoch": 2.5203252032520327,
+      "grad_norm": 0.5686483979225159,
+      "learning_rate": 1.65777536651945e-05,
+      "loss": 1.7171,
+      "num_input_tokens_seen": 926384,
+      "step": 40,
+      "train_runtime": 212.6983,
+      "train_tokens_per_second": 4355.389
+    },
+    {
+      "epoch": 2.845528455284553,
+      "grad_norm": 0.4782630205154419,
+      "learning_rate": 1.1110744174509952e-05,
+      "loss": 1.6151,
+      "num_input_tokens_seen": 1046896,
+      "step": 45,
+      "train_runtime": 240.4926,
+      "train_tokens_per_second": 4353.132
+    },
+    {
+      "epoch": 3.130081300813008,
+      "grad_norm": 0.518913745880127,
+      "learning_rate": 6.476221866126029e-06,
+      "loss": 1.4855,
+      "num_input_tokens_seen": 1151200,
+      "step": 50,
+      "train_runtime": 264.5411,
+      "train_tokens_per_second": 4351.686
+    },
+    {
+      "epoch": 3.4552845528455283,
+      "grad_norm": 0.5394554734230042,
+      "learning_rate": 2.9519683912911266e-06,
+      "loss": 1.6717,
+      "num_input_tokens_seen": 1270640,
+      "step": 55,
+      "train_runtime": 291.9391,
+      "train_tokens_per_second": 4352.414
+    },
+    {
+      "epoch": 3.7804878048780486,
+      "grad_norm": 0.7331702709197998,
+      "learning_rate": 7.492186701364007e-07,
+      "loss": 1.6515,
+      "num_input_tokens_seen": 1390528,
+      "step": 60,
+      "train_runtime": 319.5924,
+      "train_tokens_per_second": 4350.941
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 64,
+  "num_input_tokens_seen": 1471136,
+  "num_train_epochs": 4,
+  "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": 6.135952763047117e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-64/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4574205f9aa181199e18c06cfa93ed9296b0cd6271734ef844960a090a18653d
+size 6161

llamaboard_config.yaml

@@ -0,0 +1,85 @@
+top.booster: auto
+top.checkpoint_path: []
+top.finetuning_type: lora
+top.model_name: LLaVA-1.5-7B-Chat
+top.quantization_bit: none
+top.quantization_method: bnb
+top.rope_scaling: none
+top.template: llava
+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: 2048
+train.dataset:
+- sticker_labels_kfolds_train_zh_1
+train.dataset_dir: data
+train.ds_offload: false
+train.ds_stage: none
+train.enable_thinking: true
+train.extra_args: '{"optim": "adamw_torch"}'
+train.freeze_extra_modules: ''
+train.freeze_language_model: false
+train.freeze_multi_modal_projector: true
+train.freeze_trainable_layers: 2
+train.freeze_trainable_modules: all
+train.freeze_vision_tower: true
+train.galore_rank: 16
+train.galore_scale: 2
+train.galore_target: all
+train.galore_update_interval: 200
+train.gradient_accumulation_steps: 8
+train.image_max_pixels: 768*768
+train.image_min_pixels: 32*32
+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: '4'
+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.video_max_pixels: 256*256
+train.video_min_pixels: 16*16
+train.warmup_steps: 0

preprocessor_config.json

@@ -0,0 +1,28 @@
+{
+  "crop_size": {
+    "height": 336,
+    "width": 336
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "processor_class": "LlavaProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 336
+  }
+}

processor_config.json

@@ -0,0 +1,7 @@
+{
+  "image_token": "<image>",
+  "num_additional_image_tokens": 1,
+  "patch_size": 14,
+  "processor_class": "LlavaProcessor",
+  "vision_feature_select_strategy": "default"
+}

running_log.txt

@@ -0,0 +1,459 @@
+[INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file tokenizer.model from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer.model
+
+[INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file tokenizer.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer.json
+
+[INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file added_tokens.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/added_tokens.json
+
+[INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file special_tokens_map.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/special_tokens_map.json
+
+[INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file tokenizer_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer_config.json
+
+[INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2023 >> loading file chat_template.jinja from cache at None
+
+[INFO|2025-05-30 02:31:15] tokenization_utils_base.py:2299 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+[INFO|2025-05-30 02:31:17] processing_utils.py:930 >> loading configuration file processor_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/processor_config.json
+
+[WARNING|2025-05-30 02:31:17] logging.py:328 >> 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 v4.52, 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|2025-05-30 02:31:17] image_processing_base.py:380 >> loading configuration file preprocessor_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/preprocessor_config.json
+
+[WARNING|2025-05-30 02:31:17] logging.py:328 >> 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 v4.52, 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|2025-05-30 02:31:17] image_processing_base.py:433 >> Image processor CLIPImageProcessor {
+  "crop_size": {
+    "height": 336,
+    "width": 336
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "processor_class": "LlavaProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 336
+  }
+}
+
+
+[INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file tokenizer.model from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer.model
+
+[INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file tokenizer.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer.json
+
+[INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file added_tokens.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/added_tokens.json
+
+[INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file special_tokens_map.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/special_tokens_map.json
+
+[INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file tokenizer_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/tokenizer_config.json
+
+[INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2023 >> loading file chat_template.jinja from cache at None
+
+[INFO|2025-05-30 02:31:18] tokenization_utils_base.py:2299 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+[INFO|2025-05-30 02:31:19] processing_utils.py:930 >> loading configuration file processor_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/processor_config.json
+
+[INFO|2025-05-30 02:31:19] processing_utils.py:990 >> Processor LlavaProcessor:
+- image_processor: CLIPImageProcessor {
+  "crop_size": {
+    "height": 336,
+    "width": 336
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "processor_class": "LlavaProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 336
+  }
+}
+
+- tokenizer: LlamaTokenizerFast(name_or_path='llava-hf/llava-1.5-7b-hf', vocab_size=32000, model_max_length=1000000000000000019884624838656, is_fast=True, padding_side='left', truncation_side='right', special_tokens={'bos_token': '<s>', 'eos_token': '</s>', 'unk_token': '<unk>', 'pad_token': '<pad>', 'image_token': '<image>'}, clean_up_tokenization_spaces=False, added_tokens_decoder={
+	0: AddedToken("<unk>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	1: AddedToken("<s>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	2: AddedToken("</s>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	32000: AddedToken("<image>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+	32001: AddedToken("<pad>", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),
+}
+)
+
+{
+  "image_token": "<image>",
+  "num_additional_image_tokens": 1,
+  "patch_size": 14,
+  "processor_class": "LlavaProcessor",
+  "vision_feature_select_strategy": "default"
+}
+
+
+[INFO|2025-05-30 02:31:19] logging.py:143 >> Loading dataset /home/tsinghuaair/mawz/xxe_metchee/finetune-llms/llava-1.5-7b-hf-sticker-labels/kfold_output_zh/fold_1/stickers_label_train.json...
+
+[INFO|2025-05-30 02:31:24] configuration_utils.py:698 >> loading configuration file config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/config.json
+
+[INFO|2025-05-30 02:31:24] configuration_utils.py:770 >> Model config LlavaConfig {
+  "architectures": [
+    "LlavaForConditionalGeneration"
+  ],
+  "ignore_index": -100,
+  "image_seq_length": 576,
+  "image_token_index": 32000,
+  "model_type": "llava",
+  "multimodal_projector_bias": true,
+  "pad_token_id": 32001,
+  "projector_hidden_act": "gelu",
+  "text_config": {
+    "_name_or_path": "lmsys/vicuna-7b-v1.5",
+    "architectures": [
+      "LlamaForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": 11008,
+    "max_position_embeddings": 4096,
+    "mlp_bias": false,
+    "model_type": "llama",
+    "num_attention_heads": 32,
+    "num_hidden_layers": 32,
+    "num_key_value_heads": 32,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": null,
+    "rope_theta": 10000.0,
+    "torch_dtype": "float16",
+    "use_cache": true,
+    "vocab_size": 32064
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.52.1",
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "quick_gelu",
+    "hidden_size": 1024,
+    "image_size": 336,
+    "initializer_factor": 1.0,
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "layer_norm_eps": 1e-05,
+    "model_type": "clip_vision_model",
+    "num_attention_heads": 16,
+    "num_channels": 3,
+    "num_hidden_layers": 24,
+    "patch_size": 14,
+    "projection_dim": 768,
+    "vocab_size": 32000
+  },
+  "vision_feature_layer": -2,
+  "vision_feature_select_strategy": "default",
+  "vocab_size": 32064
+}
+
+
+[INFO|2025-05-30 02:31:24] logging.py:143 >> KV cache is disabled during training.
+
+[INFO|2025-05-30 02:31:24] modeling_utils.py:1149 >> loading weights file model.safetensors from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/model.safetensors.index.json
+
+[INFO|2025-05-30 02:31:24] modeling_utils.py:2239 >> Instantiating LlavaForConditionalGeneration model under default dtype torch.bfloat16.
+
+[INFO|2025-05-30 02:31:24] configuration_utils.py:1135 >> Generate config GenerationConfig {
+  "pad_token_id": 32001,
+  "use_cache": false
+}
+
+
+[INFO|2025-05-30 02:31:25] modeling_utils.py:2239 >> Instantiating CLIPVisionModel model under default dtype torch.bfloat16.
+
+[INFO|2025-05-30 02:31:25] modeling_utils.py:2239 >> Instantiating LlamaModel model under default dtype torch.bfloat16.
+
+[INFO|2025-05-30 02:31:29] modeling_utils.py:5170 >> All model checkpoint weights were used when initializing LlavaForConditionalGeneration.
+
+
+[INFO|2025-05-30 02:31:29] modeling_utils.py:5178 >> All the weights of LlavaForConditionalGeneration were initialized from the model checkpoint at llava-hf/llava-1.5-7b-hf.
+If your task is similar to the task the model of the checkpoint was trained on, you can already use LlavaForConditionalGeneration for predictions without further training.
+
+[INFO|2025-05-30 02:31:29] configuration_utils.py:1090 >> loading configuration file generation_config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/generation_config.json
+
+[INFO|2025-05-30 02:31:29] configuration_utils.py:1135 >> Generate config GenerationConfig {
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 32001
+}
+
+
+[INFO|2025-05-30 02:31:30] logging.py:143 >> Gradient checkpointing enabled.
+
+[INFO|2025-05-30 02:31:30] logging.py:143 >> Using torch SDPA for faster training and inference.
+
+[INFO|2025-05-30 02:31:30] logging.py:143 >> Upcasting trainable params to float32.
+
+[INFO|2025-05-30 02:31:30] logging.py:143 >> Fine-tuning method: LoRA
+
+[INFO|2025-05-30 02:31:30] logging.py:143 >> Found linear modules: gate_proj,q_proj,up_proj,o_proj,k_proj,v_proj,down_proj
+
+[INFO|2025-05-30 02:31:30] logging.py:143 >> Set vision model not trainable: ['vision_tower'].
+
+[INFO|2025-05-30 02:31:30] logging.py:143 >> Set multi model projector not trainable: multi_modal_projector.
+
+[INFO|2025-05-30 02:31:30] logging.py:143 >> trainable params: 19,988,480 || all params: 7,083,415,552 || trainable%: 0.2822
+
+[INFO|2025-05-30 02:31:30] trainer.py:756 >> Using auto half precision backend
+
+[INFO|2025-05-30 02:31:31] trainer.py:2409 >> ***** Running training *****
+
+[INFO|2025-05-30 02:31:31] trainer.py:2410 >>   Num examples = 489
+
+[INFO|2025-05-30 02:31:31] trainer.py:2411 >>   Num Epochs = 4
+
+[INFO|2025-05-30 02:31:31] trainer.py:2412 >>   Instantaneous batch size per device = 2
+
+[INFO|2025-05-30 02:31:31] trainer.py:2415 >>   Total train batch size (w. parallel, distributed & accumulation) = 32
+
+[INFO|2025-05-30 02:31:31] trainer.py:2416 >>   Gradient Accumulation steps = 8
+
+[INFO|2025-05-30 02:31:31] trainer.py:2417 >>   Total optimization steps = 64
+
+[INFO|2025-05-30 02:31:31] trainer.py:2418 >>   Number of trainable parameters = 19,988,480
+
+[WARNING|2025-05-30 02:31:33] logging.py:328 >> `loss_type=None` was set in the config but it is unrecognised.Using the default loss: `ForCausalLMLoss`.
+
+[WARNING|2025-05-30 02:31:33] logging.py:328 >> `loss_type=None` was set in the config but it is unrecognised.Using the default loss: `ForCausalLMLoss`.
+
+[INFO|2025-05-30 02:31:59] logging.py:143 >> {'loss': 2.2304, 'learning_rate': 4.9520e-05, 'epoch': 0.33, 'throughput': 4337.62}
+
+[INFO|2025-05-30 02:32:26] logging.py:143 >> {'loss': 2.1470, 'learning_rate': 4.7600e-05, 'epoch': 0.65, 'throughput': 4382.44}
+
+[INFO|2025-05-30 02:32:53] logging.py:143 >> {'loss': 1.9562, 'learning_rate': 4.4325e-05, 'epoch': 0.98, 'throughput': 4393.16}
+
+[INFO|2025-05-30 02:33:17] logging.py:143 >> {'loss': 1.5887, 'learning_rate': 3.9892e-05, 'epoch': 1.26, 'throughput': 4391.91}
+
+[INFO|2025-05-30 02:33:45] logging.py:143 >> {'loss': 1.7952, 'learning_rate': 3.4567e-05, 'epoch': 1.59, 'throughput': 4386.97}
+
+[INFO|2025-05-30 02:34:12] logging.py:143 >> {'loss': 1.7717, 'learning_rate': 2.8668e-05, 'epoch': 1.91, 'throughput': 4380.26}
+
+[INFO|2025-05-30 02:34:36] logging.py:143 >> {'loss': 1.5037, 'learning_rate': 2.2550e-05, 'epoch': 2.20, 'throughput': 4378.09}
+
+[INFO|2025-05-30 02:35:03] logging.py:143 >> {'loss': 1.7171, 'learning_rate': 1.6578e-05, 'epoch': 2.52, 'throughput': 4374.05}
+
+[INFO|2025-05-30 02:35:31] logging.py:143 >> {'loss': 1.6151, 'learning_rate': 1.1111e-05, 'epoch': 2.85, 'throughput': 4369.62}
+
+[INFO|2025-05-30 02:35:55] logging.py:143 >> {'loss': 1.4855, 'learning_rate': 6.4762e-06, 'epoch': 3.13, 'throughput': 4366.67}
+
+[INFO|2025-05-30 02:36:23] logging.py:143 >> {'loss': 1.6717, 'learning_rate': 2.9520e-06, 'epoch': 3.46, 'throughput': 4365.99}
+
+[INFO|2025-05-30 02:36:50] logging.py:143 >> {'loss': 1.6515, 'learning_rate': 7.4922e-07, 'epoch': 3.78, 'throughput': 4363.33}
+
+[INFO|2025-05-30 02:37:09] trainer.py:3993 >> Saving model checkpoint to saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64
+
+[INFO|2025-05-30 02:37:10] configuration_utils.py:698 >> loading configuration file config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/config.json
+
+[INFO|2025-05-30 02:37:10] configuration_utils.py:770 >> Model config LlavaConfig {
+  "architectures": [
+    "LlavaForConditionalGeneration"
+  ],
+  "ignore_index": -100,
+  "image_seq_length": 576,
+  "image_token_index": 32000,
+  "model_type": "llava",
+  "multimodal_projector_bias": true,
+  "pad_token_id": 32001,
+  "projector_hidden_act": "gelu",
+  "text_config": {
+    "_name_or_path": "lmsys/vicuna-7b-v1.5",
+    "architectures": [
+      "LlamaForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": 11008,
+    "max_position_embeddings": 4096,
+    "mlp_bias": false,
+    "model_type": "llama",
+    "num_attention_heads": 32,
+    "num_hidden_layers": 32,
+    "num_key_value_heads": 32,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": null,
+    "rope_theta": 10000.0,
+    "torch_dtype": "float16",
+    "use_cache": true,
+    "vocab_size": 32064
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.52.1",
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "quick_gelu",
+    "hidden_size": 1024,
+    "image_size": 336,
+    "initializer_factor": 1.0,
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "layer_norm_eps": 1e-05,
+    "model_type": "clip_vision_model",
+    "num_attention_heads": 16,
+    "num_channels": 3,
+    "num_hidden_layers": 24,
+    "patch_size": 14,
+    "projection_dim": 768,
+    "vocab_size": 32000
+  },
+  "vision_feature_layer": -2,
+  "vision_feature_select_strategy": "default",
+  "vocab_size": 32064
+}
+
+
+[INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2356 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/chat_template.jinja
+
+[INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2525 >> tokenizer config file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/tokenizer_config.json
+
+[INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2534 >> Special tokens file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/special_tokens_map.json
+
+[INFO|2025-05-30 02:37:10] image_processing_base.py:260 >> Image processor saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/preprocessor_config.json
+
+[INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2356 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/chat_template.jinja
+
+[INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2525 >> tokenizer config file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/tokenizer_config.json
+
+[INFO|2025-05-30 02:37:10] tokenization_utils_base.py:2534 >> Special tokens file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/special_tokens_map.json
+
+[INFO|2025-05-30 02:37:10] processing_utils.py:674 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/chat_template.jinja
+
+[INFO|2025-05-30 02:37:11] processing_utils.py:709 >> processor saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/checkpoint-64/processor_config.json
+
+[INFO|2025-05-30 02:37:11] trainer.py:2676 >> 
+
+Training completed. Do not forget to share your model on huggingface.co/models =)
+
+
+
+[INFO|2025-05-30 02:37:11] image_processing_base.py:260 >> Image processor saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/preprocessor_config.json
+
+[INFO|2025-05-30 02:37:11] tokenization_utils_base.py:2356 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/chat_template.jinja
+
+[INFO|2025-05-30 02:37:11] tokenization_utils_base.py:2525 >> tokenizer config file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/tokenizer_config.json
+
+[INFO|2025-05-30 02:37:11] tokenization_utils_base.py:2534 >> Special tokens file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/special_tokens_map.json
+
+[INFO|2025-05-30 02:37:11] processing_utils.py:674 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/chat_template.jinja
+
+[INFO|2025-05-30 02:37:11] processing_utils.py:709 >> processor saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/processor_config.json
+
+[INFO|2025-05-30 02:37:11] trainer.py:3993 >> Saving model checkpoint to saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs
+
+[INFO|2025-05-30 02:37:11] configuration_utils.py:698 >> loading configuration file config.json from cache at /home/tsinghuaair/.cache/huggingface/hub/models--llava-hf--llava-1.5-7b-hf/snapshots/6ceb2ed33cb8f107a781c431fe2e61574da69369/config.json
+
+[INFO|2025-05-30 02:37:11] configuration_utils.py:770 >> Model config LlavaConfig {
+  "architectures": [
+    "LlavaForConditionalGeneration"
+  ],
+  "ignore_index": -100,
+  "image_seq_length": 576,
+  "image_token_index": 32000,
+  "model_type": "llava",
+  "multimodal_projector_bias": true,
+  "pad_token_id": 32001,
+  "projector_hidden_act": "gelu",
+  "text_config": {
+    "_name_or_path": "lmsys/vicuna-7b-v1.5",
+    "architectures": [
+      "LlamaForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": 11008,
+    "max_position_embeddings": 4096,
+    "mlp_bias": false,
+    "model_type": "llama",
+    "num_attention_heads": 32,
+    "num_hidden_layers": 32,
+    "num_key_value_heads": 32,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": null,
+    "rope_theta": 10000.0,
+    "torch_dtype": "float16",
+    "use_cache": true,
+    "vocab_size": 32064
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.52.1",
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "quick_gelu",
+    "hidden_size": 1024,
+    "image_size": 336,
+    "initializer_factor": 1.0,
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "layer_norm_eps": 1e-05,
+    "model_type": "clip_vision_model",
+    "num_attention_heads": 16,
+    "num_channels": 3,
+    "num_hidden_layers": 24,
+    "patch_size": 14,
+    "projection_dim": 768,
+    "vocab_size": 32000
+  },
+  "vision_feature_layer": -2,
+  "vision_feature_select_strategy": "default",
+  "vocab_size": 32064
+}
+
+
+[INFO|2025-05-30 02:37:12] tokenization_utils_base.py:2356 >> chat template saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/chat_template.jinja
+
+[INFO|2025-05-30 02:37:12] tokenization_utils_base.py:2525 >> tokenizer config file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/tokenizer_config.json
+
+[INFO|2025-05-30 02:37:12] tokenization_utils_base.py:2534 >> Special tokens file saved in saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs/special_tokens_map.json
+
+[WARNING|2025-05-30 02:37:12] logging.py:148 >> No metric eval_loss to plot.
+
+[WARNING|2025-05-30 02:37:12] logging.py:148 >> No metric eval_accuracy to plot.
+
+[INFO|2025-05-30 02:37:12] modelcard.py:450 >> 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 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,65 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32000": {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {
+    "image_token": "<image>"
+  },
+  "image_token": "<image>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "padding_side": "right",
+  "processor_class": "LlavaProcessor",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "trust_remote_code": false,
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.0,
+    "num_input_tokens_seen": 1471136,
+    "total_flos": 6.135952763047117e+16,
+    "train_loss": 1.7421463951468468,
+    "train_runtime": 339.8451,
+    "train_samples_per_second": 5.756,
+    "train_steps_per_second": 0.188
+}

trainer_log.jsonl

@@ -0,0 +1,13 @@
+{"current_steps": 5, "total_steps": 64, "loss": 2.2304, "lr": 4.951963201008076e-05, "epoch": 0.3252032520325203, "percentage": 7.81, "elapsed_time": "0:00:27", "remaining_time": "0:05:23", "throughput": 4337.62, "total_tokens": 119024}
+{"current_steps": 10, "total_steps": 64, "loss": 2.147, "lr": 4.759973232808609e-05, "epoch": 0.6504065040650406, "percentage": 15.62, "elapsed_time": "0:00:54", "remaining_time": "0:04:54", "throughput": 4382.44, "total_tokens": 238880}
+{"current_steps": 15, "total_steps": 64, "loss": 1.9562, "lr": 4.4325261334068426e-05, "epoch": 0.975609756097561, "percentage": 23.44, "elapsed_time": "0:01:21", "remaining_time": "0:04:26", "throughput": 4393.16, "total_tokens": 358608}
+{"current_steps": 20, "total_steps": 64, "loss": 1.5887, "lr": 3.9892482612310836e-05, "epoch": 1.2601626016260163, "percentage": 31.25, "elapsed_time": "0:01:45", "remaining_time": "0:03:52", "throughput": 4391.91, "total_tokens": 463568}
+{"current_steps": 25, "total_steps": 64, "loss": 1.7952, "lr": 3.456708580912725e-05, "epoch": 1.5853658536585367, "percentage": 39.06, "elapsed_time": "0:02:12", "remaining_time": "0:03:27", "throughput": 4386.97, "total_tokens": 583360}
+{"current_steps": 30, "total_steps": 64, "loss": 1.7717, "lr": 2.8668261861384045e-05, "epoch": 1.910569105691057, "percentage": 46.88, "elapsed_time": "0:02:40", "remaining_time": "0:03:01", "throughput": 4380.26, "total_tokens": 702656}
+{"current_steps": 35, "total_steps": 64, "loss": 1.5037, "lr": 2.2549571491760986e-05, "epoch": 2.1951219512195124, "percentage": 54.69, "elapsed_time": "0:03:04", "remaining_time": "0:02:32", "throughput": 4378.09, "total_tokens": 807312}
+{"current_steps": 40, "total_steps": 64, "loss": 1.7171, "lr": 1.65777536651945e-05, "epoch": 2.5203252032520327, "percentage": 62.5, "elapsed_time": "0:03:31", "remaining_time": "0:02:07", "throughput": 4374.05, "total_tokens": 926384}
+{"current_steps": 45, "total_steps": 64, "loss": 1.6151, "lr": 1.1110744174509952e-05, "epoch": 2.845528455284553, "percentage": 70.31, "elapsed_time": "0:03:59", "remaining_time": "0:01:41", "throughput": 4369.62, "total_tokens": 1046896}
+{"current_steps": 50, "total_steps": 64, "loss": 1.4855, "lr": 6.476221866126029e-06, "epoch": 3.130081300813008, "percentage": 78.12, "elapsed_time": "0:04:23", "remaining_time": "0:01:13", "throughput": 4366.67, "total_tokens": 1151200}
+{"current_steps": 55, "total_steps": 64, "loss": 1.6717, "lr": 2.9519683912911266e-06, "epoch": 3.4552845528455283, "percentage": 85.94, "elapsed_time": "0:04:51", "remaining_time": "0:00:47", "throughput": 4365.99, "total_tokens": 1270640}
+{"current_steps": 60, "total_steps": 64, "loss": 1.6515, "lr": 7.492186701364007e-07, "epoch": 3.7804878048780486, "percentage": 93.75, "elapsed_time": "0:05:18", "remaining_time": "0:00:21", "throughput": 4363.33, "total_tokens": 1390528}
+{"current_steps": 64, "total_steps": 64, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "0:05:38", "remaining_time": "0:00:00", "throughput": 4340.42, "total_tokens": 1471136}

trainer_state.json

@@ -0,0 +1,164 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 64,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3252032520325203,
+      "grad_norm": 1.3156858682632446,
+      "learning_rate": 4.951963201008076e-05,
+      "loss": 2.2304,
+      "num_input_tokens_seen": 119024,
+      "step": 5,
+      "train_runtime": 28.3475,
+      "train_tokens_per_second": 4198.75
+    },
+    {
+      "epoch": 0.6504065040650406,
+      "grad_norm": 0.7530735731124878,
+      "learning_rate": 4.759973232808609e-05,
+      "loss": 2.147,
+      "num_input_tokens_seen": 238880,
+      "step": 10,
+      "train_runtime": 55.416,
+      "train_tokens_per_second": 4310.67
+    },
+    {
+      "epoch": 0.975609756097561,
+      "grad_norm": 0.6344960927963257,
+      "learning_rate": 4.4325261334068426e-05,
+      "loss": 1.9562,
+      "num_input_tokens_seen": 358608,
+      "step": 15,
+      "train_runtime": 82.5363,
+      "train_tokens_per_second": 4344.849
+    },
+    {
+      "epoch": 1.2601626016260163,
+      "grad_norm": 0.5793188214302063,
+      "learning_rate": 3.9892482612310836e-05,
+      "loss": 1.5887,
+      "num_input_tokens_seen": 463568,
+      "step": 20,
+      "train_runtime": 106.4581,
+      "train_tokens_per_second": 4354.466
+    },
+    {
+      "epoch": 1.5853658536585367,
+      "grad_norm": 0.5236935615539551,
+      "learning_rate": 3.456708580912725e-05,
+      "loss": 1.7952,
+      "num_input_tokens_seen": 583360,
+      "step": 25,
+      "train_runtime": 133.8831,
+      "train_tokens_per_second": 4357.233
+    },
+    {
+      "epoch": 1.910569105691057,
+      "grad_norm": 0.6524015665054321,
+      "learning_rate": 2.8668261861384045e-05,
+      "loss": 1.7717,
+      "num_input_tokens_seen": 702656,
+      "step": 30,
+      "train_runtime": 161.3217,
+      "train_tokens_per_second": 4355.62
+    },
+    {
+      "epoch": 2.1951219512195124,
+      "grad_norm": 0.5044527053833008,
+      "learning_rate": 2.2549571491760986e-05,
+      "loss": 1.5037,
+      "num_input_tokens_seen": 807312,
+      "step": 35,
+      "train_runtime": 185.3057,
+      "train_tokens_per_second": 4356.649
+    },
+    {
+      "epoch": 2.5203252032520327,
+      "grad_norm": 0.5686483979225159,
+      "learning_rate": 1.65777536651945e-05,
+      "loss": 1.7171,
+      "num_input_tokens_seen": 926384,
+      "step": 40,
+      "train_runtime": 212.6983,
+      "train_tokens_per_second": 4355.389
+    },
+    {
+      "epoch": 2.845528455284553,
+      "grad_norm": 0.4782630205154419,
+      "learning_rate": 1.1110744174509952e-05,
+      "loss": 1.6151,
+      "num_input_tokens_seen": 1046896,
+      "step": 45,
+      "train_runtime": 240.4926,
+      "train_tokens_per_second": 4353.132
+    },
+    {
+      "epoch": 3.130081300813008,
+      "grad_norm": 0.518913745880127,
+      "learning_rate": 6.476221866126029e-06,
+      "loss": 1.4855,
+      "num_input_tokens_seen": 1151200,
+      "step": 50,
+      "train_runtime": 264.5411,
+      "train_tokens_per_second": 4351.686
+    },
+    {
+      "epoch": 3.4552845528455283,
+      "grad_norm": 0.5394554734230042,
+      "learning_rate": 2.9519683912911266e-06,
+      "loss": 1.6717,
+      "num_input_tokens_seen": 1270640,
+      "step": 55,
+      "train_runtime": 291.9391,
+      "train_tokens_per_second": 4352.414
+    },
+    {
+      "epoch": 3.7804878048780486,
+      "grad_norm": 0.7331702709197998,
+      "learning_rate": 7.492186701364007e-07,
+      "loss": 1.6515,
+      "num_input_tokens_seen": 1390528,
+      "step": 60,
+      "train_runtime": 319.5924,
+      "train_tokens_per_second": 4350.941
+    },
+    {
+      "epoch": 4.0,
+      "num_input_tokens_seen": 1471136,
+      "step": 64,
+      "total_flos": 6.135952763047117e+16,
+      "train_loss": 1.7421463951468468,
+      "train_runtime": 339.8451,
+      "train_samples_per_second": 5.756,
+      "train_steps_per_second": 0.188
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 64,
+  "num_input_tokens_seen": 1471136,
+  "num_train_epochs": 4,
+  "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": 6.135952763047117e+16,
+  "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 sha256:4574205f9aa181199e18c06cfa93ed9296b0cd6271734ef844960a090a18653d
+size 6161

training_args.yaml

@@ -0,0 +1,40 @@
+bf16: true
+cutoff_len: 2048
+dataset: sticker_labels_kfolds_train_zh_1
+dataset_dir: data
+ddp_timeout: 180000000
+do_train: true
+enable_thinking: true
+finetuning_type: lora
+flash_attn: auto
+freeze_multi_modal_projector: true
+freeze_vision_tower: true
+gradient_accumulation_steps: 8
+image_max_pixels: 589824
+image_min_pixels: 1024
+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: llava-hf/llava-1.5-7b-hf
+num_train_epochs: 4.0
+optim: adamw_torch
+output_dir: saves/LLaVA-1.5-7B-Chat/lora/train_zh_1_fold_4_epochs
+packing: false
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+report_to: none
+save_steps: 100
+stage: sft
+template: llava
+trust_remote_code: true
+video_max_pixels: 65536
+video_min_pixels: 256
+warmup_steps: 0