Upload full training folder with all checkpoints

.gitattributes

@@ -33,3 +33,10 @@ 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
+checkpoint-220/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-230/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-240/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-250/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-260/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+final_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+lora_adapters/tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,175 @@
+---
+language:
+- vi
+- en
+license: apache-2.0
+tags:
+- llm-judge
+- training-checkpoint
+- lora
+- unsloth
+---
+
+# finetuned_4_12
+
+Full training folder backup - Toàn bộ checkpoints và models.
+
+## 📂 Cấu trúc Folder
+```
+train_
+├── lora_adapters/           # LoRA adapters
+├── README.md
+├── zero_shot_metrics.json
+└── zero_shot_results.csv
+```
+
+## 🚀 Sử Dụng
+
+### 1️⃣ Clone Repo
+```bash
+git lfs install
+git clone https://huggingface.co/ImNotTam/finetuned_4_12
+cd finetuned_4_12
+```
+
+### 2️⃣ Load LoRA Adapters (Nhẹ nhất - khuyến nghị)
+```python
+from unsloth import FastLanguageModel
+
+model, tokenizer = FastLanguageModel.from_pretrained(
+    model_name="ImNotTam/finetuned_4_12",
+    subfolder="lora_adapters",
+    max_seq_length=2048,
+    dtype=None,
+    load_in_4bit=True,
+)
+
+# Enable inference mode
+FastLanguageModel.for_inference(model)
+
+# Test
+prompt = "Đánh giá response này..."
+inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+outputs = model.generate(**inputs, max_new_tokens=256)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+
+### 3️⃣ Load Final Model
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained(
+    "ImNotTam/finetuned_4_12",
+    subfolder="final_model",
+    device_map="auto",
+    torch_dtype="auto"
+)
+tokenizer = AutoTokenizer.from_pretrained("ImNotTam/finetuned_4_12", subfolder="final_model")
+
+# Inference
+inputs = tokenizer("Your prompt", return_tensors="pt").to("cuda")
+outputs = model.generate(**inputs)
+print(tokenizer.decode(outputs[0]))
+```
+
+### 4️⃣ Resume Training từ Checkpoint
+```python
+from transformers import Trainer, TrainingArguments
+
+# Load checkpoint muốn resume
+model = AutoModelForCausalLM.from_pretrained(
+    "ImNotTam/finetuned_4_12",
+    subfolder="checkpoint-210",  # Chọn checkpoint
+    device_map="auto"
+)
+
+# Continue training
+trainer = Trainer(
+    model=model,
+    args=TrainingArguments(
+        output_dir="./continue_training",
+        # ... your training args
+    ),
+)
+trainer.train(resume_from_checkpoint=True)
+```
+
+### 5️⃣ Fine-tune Tiếp từ LoRA Adapter
+```python
+from unsloth import FastLanguageModel
+from trl import SFTTrainer
+
+# Load LoRA adapter
+model, tokenizer = FastLanguageModel.from_pretrained(
+    model_name="ImNotTam/finetuned_4_12",
+    subfolder="lora_adapters",
+    max_seq_length=2048,
+    dtype=None,
+    load_in_4bit=True,
+)
+
+# Add LoRA config để train tiếp
+model = FastLanguageModel.get_peft_model(
+    model,
+    r=16,
+    target_modules=["q_proj", "k_proj", "v_proj", "o_proj",
+                    "gate_proj", "up_proj", "down_proj"],
+    lora_alpha=16,
+    lora_dropout=0,
+    bias="none",
+    use_gradient_checkpointing="unsloth",
+)
+
+# Train với data mới
+trainer = SFTTrainer(
+    model=model,
+    tokenizer=tokenizer,
+    train_dataset=your_new_dataset,
+    # ... training args
+)
+trainer.train()
+```
+
+### 6️⃣ Xem Metrics và Results
+```python
+import json
+import pandas as pd
+
+# Load metrics
+with open("zero_shot_metrics.json", "r") as f:
+    metrics = json.load(f)
+print("📊 Metrics:", metrics)
+
+# Load results
+results = pd.read_csv("zero_shot_results.csv")
+print("\n📈 Results:")
+print(results.head())
+```
+
+## 📋 Nội Dung Repo
+
+| Folder/File | Mô tả | Kích thước |
+|-------------|-------|------------|
+| `lora_adapters/` | LoRA adapters (nhẹ) | ~50-100 MB |
+| `final_model/` | Model merged đầy đủ | ~4-8 GB |
+| `checkpoint-150/` | Training checkpoint | ~4-8 GB |
+| `checkpoint-200/` | Training checkpoint | ~4-8 GB |
+| `checkpoint-210/` | Training checkpoint | ~4-8 GB |
+| `zero_shot_metrics.json` | Evaluation metrics | <1 MB |
+| `zero_shot_results.csv` | Detailed results | <1 MB |
+
+## 💡 Khuyến Nghị
+
+- **Inference nhanh:** Dùng `lora_adapters/`
+- **Production:** Dùng `final_model/`
+- **Train tiếp:** Load `lora_adapters/` + add LoRA config
+- **Resume training:** Load checkpoint cụ thể
+
+## 📦 Requirements
+```bash
+pip install unsloth transformers torch trl
+```
+
+## 📄 License
+
+Apache 2.0

checkpoint-220/README.md

@@ -0,0 +1,210 @@
+---
+base_model: unsloth/medgemma-4b-it-unsloth-bnb-4bit
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:unsloth/medgemma-4b-it-unsloth-bnb-4bit
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+---
+
+# 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 0.18.0

checkpoint-220/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Gemma3ForConditionalGeneration",
+    "parent_library": "transformers.models.gemma3.modeling_gemma3",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/medgemma-4b-it-unsloth-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(?:.*?(?:vision|image|visual|patch|language|text).*?(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense).*?(?:k_proj|v_proj|q_proj|out_proj|fc1|fc2|o_proj|gate_proj|up_proj|down_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense)\\.(?:(?:k_proj|v_proj|q_proj|out_proj|fc1|fc2|o_proj|gate_proj|up_proj|down_proj)))",
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": true
+}

checkpoint-220/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:383e6779985c7ead01c6bfa46928f6b10f7acf13e095c9416f48b1d37155a748
+size 308107928

checkpoint-220/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

checkpoint-220/chat_template.jinja

@@ -0,0 +1,47 @@
+{{ bos_token }}
+{%- if messages[0]['role'] == 'system' -%}
+    {%- if messages[0]['content'] is string -%}
+        {%- set first_user_prefix = messages[0]['content'] + '
+
+' -%}
+    {%- else -%}
+        {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
+
+' -%}
+    {%- endif -%}
+    {%- set loop_messages = messages[1:] -%}
+{%- else -%}
+    {%- set first_user_prefix = "" -%}
+    {%- set loop_messages = messages -%}
+{%- endif -%}
+{%- for message in loop_messages -%}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
+        {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
+    {%- endif -%}
+    {%- if (message['role'] == 'assistant') -%}
+        {%- set role = "model" -%}
+    {%- else -%}
+        {%- set role = message['role'] -%}
+    {%- endif -%}
+    {{ '<start_of_turn>' + role + '
+' + (first_user_prefix if loop.first else "") }}
+    {%- if message['content'] is string -%}
+        {{ message['content'] | trim }}
+    {%- elif message['content'] is iterable -%}
+        {%- for item in message['content'] -%}
+            {%- if item['type'] == 'image' -%}
+                {{ '<start_of_image>' }}
+            {%- elif item['type'] == 'text' -%}
+                {{ item['text'] | trim }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{ raise_exception("Invalid content type") }}
+    {%- endif -%}
+    {{ '<end_of_turn>
+' }}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    {{'<start_of_turn>model
+'}}
+{%- endif -%}

checkpoint-220/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4dd9ee185873bdd05083b4b486621d58a40bfefe9c0d53e2ca1fbd34c4e06f01
+size 157331237

checkpoint-220/preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_pan_and_scan": null,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "Gemma3ImageProcessor",
+  "image_seq_length": 256,
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "pan_and_scan_max_num_crops": null,
+  "pan_and_scan_min_crop_size": null,
+  "pan_and_scan_min_ratio_to_activate": null,
+  "processor_class": "Gemma3Processor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 896,
+    "width": 896
+  }
+}

checkpoint-220/processor_config.json

@@ -0,0 +1,4 @@
+{
+  "image_seq_length": 256,
+  "processor_class": "Gemma3Processor"
+}

checkpoint-220/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f58f67ef5f5176e222513e3a025ff13130dac72dee0647315ab995c26de4a4bd
+size 14645

checkpoint-220/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8cd0f7e390bbc1ed624f589a4721faeebacc227b89e31ec424dd12abd261a626
+size 1465

checkpoint-220/special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<end_of_turn>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "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-220/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7666402c0617d170e6b0a985b3130c3fb0795393aa0970600994a5d9aae12351
+size 33384822

checkpoint-220/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

checkpoint-220/trainer_state.json

@@ -0,0 +1,527 @@
+{
+  "best_global_step": 220,
+  "best_metric": 0.010845971293747425,
+  "best_model_checkpoint": "/teamspace/studios/this_studio/DATN/output/medgemma_finetuned/checkpoint-220",
+  "epoch": 0.8560311284046692,
+  "eval_steps": 10,
+  "global_step": 220,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.019455252918287938,
+      "grad_norm": 3.7302613258361816,
+      "learning_rate": 1.777777777777778e-06,
+      "loss": 0.8239,
+      "step": 5
+    },
+    {
+      "epoch": 0.038910505836575876,
+      "grad_norm": 3.058382034301758,
+      "learning_rate": 4e-06,
+      "loss": 0.7964,
+      "step": 10
+    },
+    {
+      "epoch": 0.038910505836575876,
+      "eval_loss": 0.7571244239807129,
+      "eval_runtime": 152.7626,
+      "eval_samples_per_second": 3.339,
+      "eval_steps_per_second": 0.838,
+      "step": 10
+    },
+    {
+      "epoch": 0.058365758754863814,
+      "grad_norm": 1.8304738998413086,
+      "learning_rate": 6.222222222222222e-06,
+      "loss": 0.6943,
+      "step": 15
+    },
+    {
+      "epoch": 0.07782101167315175,
+      "grad_norm": 1.3162634372711182,
+      "learning_rate": 8.444444444444446e-06,
+      "loss": 0.5569,
+      "step": 20
+    },
+    {
+      "epoch": 0.07782101167315175,
+      "eval_loss": 0.5357815623283386,
+      "eval_runtime": 136.4123,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 20
+    },
+    {
+      "epoch": 0.09727626459143969,
+      "grad_norm": 1.1865513324737549,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.4222,
+      "step": 25
+    },
+    {
+      "epoch": 0.11673151750972763,
+      "grad_norm": 1.2043355703353882,
+      "learning_rate": 1.2888888888888889e-05,
+      "loss": 0.2831,
+      "step": 30
+    },
+    {
+      "epoch": 0.11673151750972763,
+      "eval_loss": 0.28264933824539185,
+      "eval_runtime": 136.5183,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 30
+    },
+    {
+      "epoch": 0.13618677042801555,
+      "grad_norm": 0.9503483772277832,
+      "learning_rate": 1.511111111111111e-05,
+      "loss": 0.1405,
+      "step": 35
+    },
+    {
+      "epoch": 0.1556420233463035,
+      "grad_norm": 0.34179583191871643,
+      "learning_rate": 1.7333333333333332e-05,
+      "loss": 0.0515,
+      "step": 40
+    },
+    {
+      "epoch": 0.1556420233463035,
+      "eval_loss": 0.05298379808664322,
+      "eval_runtime": 136.391,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 40
+    },
+    {
+      "epoch": 0.17509727626459143,
+      "grad_norm": 0.24818392097949982,
+      "learning_rate": 1.9555555555555557e-05,
+      "loss": 0.0302,
+      "step": 45
+    },
+    {
+      "epoch": 0.19455252918287938,
+      "grad_norm": 0.19131316244602203,
+      "learning_rate": 2.177777777777778e-05,
+      "loss": 0.0221,
+      "step": 50
+    },
+    {
+      "epoch": 0.19455252918287938,
+      "eval_loss": 0.024967821314930916,
+      "eval_runtime": 136.5245,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 50
+    },
+    {
+      "epoch": 0.2140077821011673,
+      "grad_norm": 2.454702377319336,
+      "learning_rate": 2.4e-05,
+      "loss": 0.0169,
+      "step": 55
+    },
+    {
+      "epoch": 0.23346303501945526,
+      "grad_norm": 0.12364993244409561,
+      "learning_rate": 2.6222222222222226e-05,
+      "loss": 0.0164,
+      "step": 60
+    },
+    {
+      "epoch": 0.23346303501945526,
+      "eval_loss": 0.021661706268787384,
+      "eval_runtime": 136.8436,
+      "eval_samples_per_second": 3.727,
+      "eval_steps_per_second": 0.935,
+      "step": 60
+    },
+    {
+      "epoch": 0.2529182879377432,
+      "grad_norm": 0.14085163176059723,
+      "learning_rate": 2.8444444444444447e-05,
+      "loss": 0.0131,
+      "step": 65
+    },
+    {
+      "epoch": 0.2723735408560311,
+      "grad_norm": 0.15322668850421906,
+      "learning_rate": 3.066666666666666e-05,
+      "loss": 0.0089,
+      "step": 70
+    },
+    {
+      "epoch": 0.2723735408560311,
+      "eval_loss": 0.016315119341015816,
+      "eval_runtime": 136.2077,
+      "eval_samples_per_second": 3.744,
+      "eval_steps_per_second": 0.94,
+      "step": 70
+    },
+    {
+      "epoch": 0.2918287937743191,
+      "grad_norm": 0.08343034237623215,
+      "learning_rate": 3.288888888888889e-05,
+      "loss": 0.0076,
+      "step": 75
+    },
+    {
+      "epoch": 0.311284046692607,
+      "grad_norm": 0.11078440397977829,
+      "learning_rate": 3.511111111111111e-05,
+      "loss": 0.008,
+      "step": 80
+    },
+    {
+      "epoch": 0.311284046692607,
+      "eval_loss": 0.015718888491392136,
+      "eval_runtime": 136.7712,
+      "eval_samples_per_second": 3.729,
+      "eval_steps_per_second": 0.936,
+      "step": 80
+    },
+    {
+      "epoch": 0.33073929961089493,
+      "grad_norm": 0.08361168950796127,
+      "learning_rate": 3.733333333333334e-05,
+      "loss": 0.008,
+      "step": 85
+    },
+    {
+      "epoch": 0.35019455252918286,
+      "grad_norm": 0.06539439409971237,
+      "learning_rate": 3.9555555555555556e-05,
+      "loss": 0.0083,
+      "step": 90
+    },
+    {
+      "epoch": 0.35019455252918286,
+      "eval_loss": 0.015910081565380096,
+      "eval_runtime": 136.5187,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 90
+    },
+    {
+      "epoch": 0.36964980544747084,
+      "grad_norm": 0.14973388612270355,
+      "learning_rate": 4.177777777777778e-05,
+      "loss": 0.0085,
+      "step": 95
+    },
+    {
+      "epoch": 0.38910505836575876,
+      "grad_norm": 0.08519362658262253,
+      "learning_rate": 4.4e-05,
+      "loss": 0.0077,
+      "step": 100
+    },
+    {
+      "epoch": 0.38910505836575876,
+      "eval_loss": 0.01615685597062111,
+      "eval_runtime": 136.4452,
+      "eval_samples_per_second": 3.738,
+      "eval_steps_per_second": 0.938,
+      "step": 100
+    },
+    {
+      "epoch": 0.4085603112840467,
+      "grad_norm": 0.05565109848976135,
+      "learning_rate": 4.6222222222222224e-05,
+      "loss": 0.0085,
+      "step": 105
+    },
+    {
+      "epoch": 0.4280155642023346,
+      "grad_norm": 0.07286959886550903,
+      "learning_rate": 4.844444444444445e-05,
+      "loss": 0.0082,
+      "step": 110
+    },
+    {
+      "epoch": 0.4280155642023346,
+      "eval_loss": 0.015698084607720375,
+      "eval_runtime": 136.3832,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.939,
+      "step": 110
+    },
+    {
+      "epoch": 0.4474708171206226,
+      "grad_norm": 0.13329896330833435,
+      "learning_rate": 5.066666666666667e-05,
+      "loss": 0.0085,
+      "step": 115
+    },
+    {
+      "epoch": 0.4669260700389105,
+      "grad_norm": 0.04628467932343483,
+      "learning_rate": 5.288888888888889e-05,
+      "loss": 0.0075,
+      "step": 120
+    },
+    {
+      "epoch": 0.4669260700389105,
+      "eval_loss": 0.015623296611011028,
+      "eval_runtime": 136.0422,
+      "eval_samples_per_second": 3.749,
+      "eval_steps_per_second": 0.941,
+      "step": 120
+    },
+    {
+      "epoch": 0.48638132295719844,
+      "grad_norm": 0.058520544320344925,
+      "learning_rate": 5.511111111111111e-05,
+      "loss": 0.0079,
+      "step": 125
+    },
+    {
+      "epoch": 0.5058365758754864,
+      "grad_norm": 0.06411632895469666,
+      "learning_rate": 5.7333333333333336e-05,
+      "loss": 0.0076,
+      "step": 130
+    },
+    {
+      "epoch": 0.5058365758754864,
+      "eval_loss": 0.014402530156075954,
+      "eval_runtime": 135.5623,
+      "eval_samples_per_second": 3.762,
+      "eval_steps_per_second": 0.944,
+      "step": 130
+    },
+    {
+      "epoch": 0.5252918287937743,
+      "grad_norm": 0.044081032276153564,
+      "learning_rate": 5.9555555555555554e-05,
+      "loss": 0.0072,
+      "step": 135
+    },
+    {
+      "epoch": 0.5447470817120622,
+      "grad_norm": 0.04867592826485634,
+      "learning_rate": 6.177777777777779e-05,
+      "loss": 0.0077,
+      "step": 140
+    },
+    {
+      "epoch": 0.5447470817120622,
+      "eval_loss": 0.012970623560249805,
+      "eval_runtime": 137.0137,
+      "eval_samples_per_second": 3.722,
+      "eval_steps_per_second": 0.934,
+      "step": 140
+    },
+    {
+      "epoch": 0.5642023346303502,
+      "grad_norm": 0.044633813202381134,
+      "learning_rate": 6.4e-05,
+      "loss": 0.0077,
+      "step": 145
+    },
+    {
+      "epoch": 0.5836575875486382,
+      "grad_norm": 0.052950419485569,
+      "learning_rate": 6.622222222222222e-05,
+      "loss": 0.008,
+      "step": 150
+    },
+    {
+      "epoch": 0.5836575875486382,
+      "eval_loss": 0.012441293336451054,
+      "eval_runtime": 136.133,
+      "eval_samples_per_second": 3.746,
+      "eval_steps_per_second": 0.94,
+      "step": 150
+    },
+    {
+      "epoch": 0.603112840466926,
+      "grad_norm": 0.039904553443193436,
+      "learning_rate": 6.844444444444445e-05,
+      "loss": 0.0078,
+      "step": 155
+    },
+    {
+      "epoch": 0.622568093385214,
+      "grad_norm": 0.05680263414978981,
+      "learning_rate": 7.066666666666667e-05,
+      "loss": 0.0074,
+      "step": 160
+    },
+    {
+      "epoch": 0.622568093385214,
+      "eval_loss": 0.01192025002092123,
+      "eval_runtime": 136.5582,
+      "eval_samples_per_second": 3.735,
+      "eval_steps_per_second": 0.937,
+      "step": 160
+    },
+    {
+      "epoch": 0.642023346303502,
+      "grad_norm": 0.05537933111190796,
+      "learning_rate": 7.288888888888888e-05,
+      "loss": 0.0076,
+      "step": 165
+    },
+    {
+      "epoch": 0.6614785992217899,
+      "grad_norm": 0.04935755953192711,
+      "learning_rate": 7.511111111111111e-05,
+      "loss": 0.0077,
+      "step": 170
+    },
+    {
+      "epoch": 0.6614785992217899,
+      "eval_loss": 0.012302271090447903,
+      "eval_runtime": 136.6757,
+      "eval_samples_per_second": 3.731,
+      "eval_steps_per_second": 0.937,
+      "step": 170
+    },
+    {
+      "epoch": 0.6809338521400778,
+      "grad_norm": 0.05575108528137207,
+      "learning_rate": 7.733333333333333e-05,
+      "loss": 0.0081,
+      "step": 175
+    },
+    {
+      "epoch": 0.7003891050583657,
+      "grad_norm": 0.0551481656730175,
+      "learning_rate": 7.955555555555556e-05,
+      "loss": 0.0081,
+      "step": 180
+    },
+    {
+      "epoch": 0.7003891050583657,
+      "eval_loss": 0.011542496271431446,
+      "eval_runtime": 136.4051,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 180
+    },
+    {
+      "epoch": 0.7198443579766537,
+      "grad_norm": 0.04738597571849823,
+      "learning_rate": 8.177777777777778e-05,
+      "loss": 0.0076,
+      "step": 185
+    },
+    {
+      "epoch": 0.7392996108949417,
+      "grad_norm": 0.029748599976301193,
+      "learning_rate": 8.4e-05,
+      "loss": 0.0073,
+      "step": 190
+    },
+    {
+      "epoch": 0.7392996108949417,
+      "eval_loss": 0.012059729546308517,
+      "eval_runtime": 135.8659,
+      "eval_samples_per_second": 3.754,
+      "eval_steps_per_second": 0.942,
+      "step": 190
+    },
+    {
+      "epoch": 0.7587548638132295,
+      "grad_norm": 0.03995237499475479,
+      "learning_rate": 8.622222222222223e-05,
+      "loss": 0.0077,
+      "step": 195
+    },
+    {
+      "epoch": 0.7782101167315175,
+      "grad_norm": 0.02774854749441147,
+      "learning_rate": 8.844444444444445e-05,
+      "loss": 0.0075,
+      "step": 200
+    },
+    {
+      "epoch": 0.7782101167315175,
+      "eval_loss": 0.011773883365094662,
+      "eval_runtime": 136.231,
+      "eval_samples_per_second": 3.744,
+      "eval_steps_per_second": 0.94,
+      "step": 200
+    },
+    {
+      "epoch": 0.7976653696498055,
+      "grad_norm": 0.026570243760943413,
+      "learning_rate": 9.066666666666667e-05,
+      "loss": 0.0072,
+      "step": 205
+    },
+    {
+      "epoch": 0.8171206225680934,
+      "grad_norm": 0.047289494425058365,
+      "learning_rate": 9.288888888888888e-05,
+      "loss": 0.0074,
+      "step": 210
+    },
+    {
+      "epoch": 0.8171206225680934,
+      "eval_loss": 0.011569861322641373,
+      "eval_runtime": 136.4761,
+      "eval_samples_per_second": 3.737,
+      "eval_steps_per_second": 0.938,
+      "step": 210
+    },
+    {
+      "epoch": 0.8365758754863813,
+      "grad_norm": 0.036366503685712814,
+      "learning_rate": 9.511111111111112e-05,
+      "loss": 0.007,
+      "step": 215
+    },
+    {
+      "epoch": 0.8560311284046692,
+      "grad_norm": 0.07178617268800735,
+      "learning_rate": 9.733333333333333e-05,
+      "loss": 0.0073,
+      "step": 220
+    },
+    {
+      "epoch": 0.8560311284046692,
+      "eval_loss": 0.010845971293747425,
+      "eval_runtime": 136.7794,
+      "eval_samples_per_second": 3.729,
+      "eval_steps_per_second": 0.936,
+      "step": 220
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1799,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 10,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 30,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 8
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.2998780541949824e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-220/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfc74119a8e3587a81c422ff23316abe5d974ec5c66265926247177773d932ad
+size 6417

checkpoint-230/README.md

@@ -0,0 +1,210 @@
+---
+base_model: unsloth/medgemma-4b-it-unsloth-bnb-4bit
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:unsloth/medgemma-4b-it-unsloth-bnb-4bit
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+---
+
+# 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 0.18.0

checkpoint-230/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Gemma3ForConditionalGeneration",
+    "parent_library": "transformers.models.gemma3.modeling_gemma3",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/medgemma-4b-it-unsloth-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(?:.*?(?:vision|image|visual|patch|language|text).*?(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense).*?(?:k_proj|v_proj|q_proj|out_proj|fc1|fc2|o_proj|gate_proj|up_proj|down_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense)\\.(?:(?:k_proj|v_proj|q_proj|out_proj|fc1|fc2|o_proj|gate_proj|up_proj|down_proj)))",
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": true
+}

checkpoint-230/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e828b2ee1f0e5e2cf604e123a55d06e0291c14c5ee70d0b0081aad9b22badae
+size 308107928

checkpoint-230/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

checkpoint-230/chat_template.jinja

@@ -0,0 +1,47 @@
+{{ bos_token }}
+{%- if messages[0]['role'] == 'system' -%}
+    {%- if messages[0]['content'] is string -%}
+        {%- set first_user_prefix = messages[0]['content'] + '
+
+' -%}
+    {%- else -%}
+        {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
+
+' -%}
+    {%- endif -%}
+    {%- set loop_messages = messages[1:] -%}
+{%- else -%}
+    {%- set first_user_prefix = "" -%}
+    {%- set loop_messages = messages -%}
+{%- endif -%}
+{%- for message in loop_messages -%}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
+        {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
+    {%- endif -%}
+    {%- if (message['role'] == 'assistant') -%}
+        {%- set role = "model" -%}
+    {%- else -%}
+        {%- set role = message['role'] -%}
+    {%- endif -%}
+    {{ '<start_of_turn>' + role + '
+' + (first_user_prefix if loop.first else "") }}
+    {%- if message['content'] is string -%}
+        {{ message['content'] | trim }}
+    {%- elif message['content'] is iterable -%}
+        {%- for item in message['content'] -%}
+            {%- if item['type'] == 'image' -%}
+                {{ '<start_of_image>' }}
+            {%- elif item['type'] == 'text' -%}
+                {{ item['text'] | trim }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{ raise_exception("Invalid content type") }}
+    {%- endif -%}
+    {{ '<end_of_turn>
+' }}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    {{'<start_of_turn>model
+'}}
+{%- endif -%}

checkpoint-230/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d1a1778c960f9b7620ff897419cc14a8bc8817077bb53c88a5b3fc801d6d803
+size 157331237

checkpoint-230/preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_pan_and_scan": null,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "Gemma3ImageProcessor",
+  "image_seq_length": 256,
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "pan_and_scan_max_num_crops": null,
+  "pan_and_scan_min_crop_size": null,
+  "pan_and_scan_min_ratio_to_activate": null,
+  "processor_class": "Gemma3Processor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 896,
+    "width": 896
+  }
+}

checkpoint-230/processor_config.json

@@ -0,0 +1,4 @@
+{
+  "image_seq_length": 256,
+  "processor_class": "Gemma3Processor"
+}

checkpoint-230/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc08282850ca3ef384c91a39dc5e90350a02073687b7716c9c62dd8e3b07669f
+size 14645

checkpoint-230/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a77a2f6f325b52e5dc17fbd21b1e476f64e9338f4b0973c0706b0eb0105d5c51
+size 1465

checkpoint-230/special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<end_of_turn>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "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-230/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7666402c0617d170e6b0a985b3130c3fb0795393aa0970600994a5d9aae12351
+size 33384822

checkpoint-230/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

checkpoint-230/trainer_state.json

@@ -0,0 +1,549 @@
+{
+  "best_global_step": 220,
+  "best_metric": 0.010845971293747425,
+  "best_model_checkpoint": "/teamspace/studios/this_studio/DATN/output/medgemma_finetuned/checkpoint-220",
+  "epoch": 0.8949416342412452,
+  "eval_steps": 10,
+  "global_step": 230,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.019455252918287938,
+      "grad_norm": 3.7302613258361816,
+      "learning_rate": 1.777777777777778e-06,
+      "loss": 0.8239,
+      "step": 5
+    },
+    {
+      "epoch": 0.038910505836575876,
+      "grad_norm": 3.058382034301758,
+      "learning_rate": 4e-06,
+      "loss": 0.7964,
+      "step": 10
+    },
+    {
+      "epoch": 0.038910505836575876,
+      "eval_loss": 0.7571244239807129,
+      "eval_runtime": 152.7626,
+      "eval_samples_per_second": 3.339,
+      "eval_steps_per_second": 0.838,
+      "step": 10
+    },
+    {
+      "epoch": 0.058365758754863814,
+      "grad_norm": 1.8304738998413086,
+      "learning_rate": 6.222222222222222e-06,
+      "loss": 0.6943,
+      "step": 15
+    },
+    {
+      "epoch": 0.07782101167315175,
+      "grad_norm": 1.3162634372711182,
+      "learning_rate": 8.444444444444446e-06,
+      "loss": 0.5569,
+      "step": 20
+    },
+    {
+      "epoch": 0.07782101167315175,
+      "eval_loss": 0.5357815623283386,
+      "eval_runtime": 136.4123,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 20
+    },
+    {
+      "epoch": 0.09727626459143969,
+      "grad_norm": 1.1865513324737549,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.4222,
+      "step": 25
+    },
+    {
+      "epoch": 0.11673151750972763,
+      "grad_norm": 1.2043355703353882,
+      "learning_rate": 1.2888888888888889e-05,
+      "loss": 0.2831,
+      "step": 30
+    },
+    {
+      "epoch": 0.11673151750972763,
+      "eval_loss": 0.28264933824539185,
+      "eval_runtime": 136.5183,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 30
+    },
+    {
+      "epoch": 0.13618677042801555,
+      "grad_norm": 0.9503483772277832,
+      "learning_rate": 1.511111111111111e-05,
+      "loss": 0.1405,
+      "step": 35
+    },
+    {
+      "epoch": 0.1556420233463035,
+      "grad_norm": 0.34179583191871643,
+      "learning_rate": 1.7333333333333332e-05,
+      "loss": 0.0515,
+      "step": 40
+    },
+    {
+      "epoch": 0.1556420233463035,
+      "eval_loss": 0.05298379808664322,
+      "eval_runtime": 136.391,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 40
+    },
+    {
+      "epoch": 0.17509727626459143,
+      "grad_norm": 0.24818392097949982,
+      "learning_rate": 1.9555555555555557e-05,
+      "loss": 0.0302,
+      "step": 45
+    },
+    {
+      "epoch": 0.19455252918287938,
+      "grad_norm": 0.19131316244602203,
+      "learning_rate": 2.177777777777778e-05,
+      "loss": 0.0221,
+      "step": 50
+    },
+    {
+      "epoch": 0.19455252918287938,
+      "eval_loss": 0.024967821314930916,
+      "eval_runtime": 136.5245,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 50
+    },
+    {
+      "epoch": 0.2140077821011673,
+      "grad_norm": 2.454702377319336,
+      "learning_rate": 2.4e-05,
+      "loss": 0.0169,
+      "step": 55
+    },
+    {
+      "epoch": 0.23346303501945526,
+      "grad_norm": 0.12364993244409561,
+      "learning_rate": 2.6222222222222226e-05,
+      "loss": 0.0164,
+      "step": 60
+    },
+    {
+      "epoch": 0.23346303501945526,
+      "eval_loss": 0.021661706268787384,
+      "eval_runtime": 136.8436,
+      "eval_samples_per_second": 3.727,
+      "eval_steps_per_second": 0.935,
+      "step": 60
+    },
+    {
+      "epoch": 0.2529182879377432,
+      "grad_norm": 0.14085163176059723,
+      "learning_rate": 2.8444444444444447e-05,
+      "loss": 0.0131,
+      "step": 65
+    },
+    {
+      "epoch": 0.2723735408560311,
+      "grad_norm": 0.15322668850421906,
+      "learning_rate": 3.066666666666666e-05,
+      "loss": 0.0089,
+      "step": 70
+    },
+    {
+      "epoch": 0.2723735408560311,
+      "eval_loss": 0.016315119341015816,
+      "eval_runtime": 136.2077,
+      "eval_samples_per_second": 3.744,
+      "eval_steps_per_second": 0.94,
+      "step": 70
+    },
+    {
+      "epoch": 0.2918287937743191,
+      "grad_norm": 0.08343034237623215,
+      "learning_rate": 3.288888888888889e-05,
+      "loss": 0.0076,
+      "step": 75
+    },
+    {
+      "epoch": 0.311284046692607,
+      "grad_norm": 0.11078440397977829,
+      "learning_rate": 3.511111111111111e-05,
+      "loss": 0.008,
+      "step": 80
+    },
+    {
+      "epoch": 0.311284046692607,
+      "eval_loss": 0.015718888491392136,
+      "eval_runtime": 136.7712,
+      "eval_samples_per_second": 3.729,
+      "eval_steps_per_second": 0.936,
+      "step": 80
+    },
+    {
+      "epoch": 0.33073929961089493,
+      "grad_norm": 0.08361168950796127,
+      "learning_rate": 3.733333333333334e-05,
+      "loss": 0.008,
+      "step": 85
+    },
+    {
+      "epoch": 0.35019455252918286,
+      "grad_norm": 0.06539439409971237,
+      "learning_rate": 3.9555555555555556e-05,
+      "loss": 0.0083,
+      "step": 90
+    },
+    {
+      "epoch": 0.35019455252918286,
+      "eval_loss": 0.015910081565380096,
+      "eval_runtime": 136.5187,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 90
+    },
+    {
+      "epoch": 0.36964980544747084,
+      "grad_norm": 0.14973388612270355,
+      "learning_rate": 4.177777777777778e-05,
+      "loss": 0.0085,
+      "step": 95
+    },
+    {
+      "epoch": 0.38910505836575876,
+      "grad_norm": 0.08519362658262253,
+      "learning_rate": 4.4e-05,
+      "loss": 0.0077,
+      "step": 100
+    },
+    {
+      "epoch": 0.38910505836575876,
+      "eval_loss": 0.01615685597062111,
+      "eval_runtime": 136.4452,
+      "eval_samples_per_second": 3.738,
+      "eval_steps_per_second": 0.938,
+      "step": 100
+    },
+    {
+      "epoch": 0.4085603112840467,
+      "grad_norm": 0.05565109848976135,
+      "learning_rate": 4.6222222222222224e-05,
+      "loss": 0.0085,
+      "step": 105
+    },
+    {
+      "epoch": 0.4280155642023346,
+      "grad_norm": 0.07286959886550903,
+      "learning_rate": 4.844444444444445e-05,
+      "loss": 0.0082,
+      "step": 110
+    },
+    {
+      "epoch": 0.4280155642023346,
+      "eval_loss": 0.015698084607720375,
+      "eval_runtime": 136.3832,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.939,
+      "step": 110
+    },
+    {
+      "epoch": 0.4474708171206226,
+      "grad_norm": 0.13329896330833435,
+      "learning_rate": 5.066666666666667e-05,
+      "loss": 0.0085,
+      "step": 115
+    },
+    {
+      "epoch": 0.4669260700389105,
+      "grad_norm": 0.04628467932343483,
+      "learning_rate": 5.288888888888889e-05,
+      "loss": 0.0075,
+      "step": 120
+    },
+    {
+      "epoch": 0.4669260700389105,
+      "eval_loss": 0.015623296611011028,
+      "eval_runtime": 136.0422,
+      "eval_samples_per_second": 3.749,
+      "eval_steps_per_second": 0.941,
+      "step": 120
+    },
+    {
+      "epoch": 0.48638132295719844,
+      "grad_norm": 0.058520544320344925,
+      "learning_rate": 5.511111111111111e-05,
+      "loss": 0.0079,
+      "step": 125
+    },
+    {
+      "epoch": 0.5058365758754864,
+      "grad_norm": 0.06411632895469666,
+      "learning_rate": 5.7333333333333336e-05,
+      "loss": 0.0076,
+      "step": 130
+    },
+    {
+      "epoch": 0.5058365758754864,
+      "eval_loss": 0.014402530156075954,
+      "eval_runtime": 135.5623,
+      "eval_samples_per_second": 3.762,
+      "eval_steps_per_second": 0.944,
+      "step": 130
+    },
+    {
+      "epoch": 0.5252918287937743,
+      "grad_norm": 0.044081032276153564,
+      "learning_rate": 5.9555555555555554e-05,
+      "loss": 0.0072,
+      "step": 135
+    },
+    {
+      "epoch": 0.5447470817120622,
+      "grad_norm": 0.04867592826485634,
+      "learning_rate": 6.177777777777779e-05,
+      "loss": 0.0077,
+      "step": 140
+    },
+    {
+      "epoch": 0.5447470817120622,
+      "eval_loss": 0.012970623560249805,
+      "eval_runtime": 137.0137,
+      "eval_samples_per_second": 3.722,
+      "eval_steps_per_second": 0.934,
+      "step": 140
+    },
+    {
+      "epoch": 0.5642023346303502,
+      "grad_norm": 0.044633813202381134,
+      "learning_rate": 6.4e-05,
+      "loss": 0.0077,
+      "step": 145
+    },
+    {
+      "epoch": 0.5836575875486382,
+      "grad_norm": 0.052950419485569,
+      "learning_rate": 6.622222222222222e-05,
+      "loss": 0.008,
+      "step": 150
+    },
+    {
+      "epoch": 0.5836575875486382,
+      "eval_loss": 0.012441293336451054,
+      "eval_runtime": 136.133,
+      "eval_samples_per_second": 3.746,
+      "eval_steps_per_second": 0.94,
+      "step": 150
+    },
+    {
+      "epoch": 0.603112840466926,
+      "grad_norm": 0.039904553443193436,
+      "learning_rate": 6.844444444444445e-05,
+      "loss": 0.0078,
+      "step": 155
+    },
+    {
+      "epoch": 0.622568093385214,
+      "grad_norm": 0.05680263414978981,
+      "learning_rate": 7.066666666666667e-05,
+      "loss": 0.0074,
+      "step": 160
+    },
+    {
+      "epoch": 0.622568093385214,
+      "eval_loss": 0.01192025002092123,
+      "eval_runtime": 136.5582,
+      "eval_samples_per_second": 3.735,
+      "eval_steps_per_second": 0.937,
+      "step": 160
+    },
+    {
+      "epoch": 0.642023346303502,
+      "grad_norm": 0.05537933111190796,
+      "learning_rate": 7.288888888888888e-05,
+      "loss": 0.0076,
+      "step": 165
+    },
+    {
+      "epoch": 0.6614785992217899,
+      "grad_norm": 0.04935755953192711,
+      "learning_rate": 7.511111111111111e-05,
+      "loss": 0.0077,
+      "step": 170
+    },
+    {
+      "epoch": 0.6614785992217899,
+      "eval_loss": 0.012302271090447903,
+      "eval_runtime": 136.6757,
+      "eval_samples_per_second": 3.731,
+      "eval_steps_per_second": 0.937,
+      "step": 170
+    },
+    {
+      "epoch": 0.6809338521400778,
+      "grad_norm": 0.05575108528137207,
+      "learning_rate": 7.733333333333333e-05,
+      "loss": 0.0081,
+      "step": 175
+    },
+    {
+      "epoch": 0.7003891050583657,
+      "grad_norm": 0.0551481656730175,
+      "learning_rate": 7.955555555555556e-05,
+      "loss": 0.0081,
+      "step": 180
+    },
+    {
+      "epoch": 0.7003891050583657,
+      "eval_loss": 0.011542496271431446,
+      "eval_runtime": 136.4051,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 180
+    },
+    {
+      "epoch": 0.7198443579766537,
+      "grad_norm": 0.04738597571849823,
+      "learning_rate": 8.177777777777778e-05,
+      "loss": 0.0076,
+      "step": 185
+    },
+    {
+      "epoch": 0.7392996108949417,
+      "grad_norm": 0.029748599976301193,
+      "learning_rate": 8.4e-05,
+      "loss": 0.0073,
+      "step": 190
+    },
+    {
+      "epoch": 0.7392996108949417,
+      "eval_loss": 0.012059729546308517,
+      "eval_runtime": 135.8659,
+      "eval_samples_per_second": 3.754,
+      "eval_steps_per_second": 0.942,
+      "step": 190
+    },
+    {
+      "epoch": 0.7587548638132295,
+      "grad_norm": 0.03995237499475479,
+      "learning_rate": 8.622222222222223e-05,
+      "loss": 0.0077,
+      "step": 195
+    },
+    {
+      "epoch": 0.7782101167315175,
+      "grad_norm": 0.02774854749441147,
+      "learning_rate": 8.844444444444445e-05,
+      "loss": 0.0075,
+      "step": 200
+    },
+    {
+      "epoch": 0.7782101167315175,
+      "eval_loss": 0.011773883365094662,
+      "eval_runtime": 136.231,
+      "eval_samples_per_second": 3.744,
+      "eval_steps_per_second": 0.94,
+      "step": 200
+    },
+    {
+      "epoch": 0.7976653696498055,
+      "grad_norm": 0.026570243760943413,
+      "learning_rate": 9.066666666666667e-05,
+      "loss": 0.0072,
+      "step": 205
+    },
+    {
+      "epoch": 0.8171206225680934,
+      "grad_norm": 0.047289494425058365,
+      "learning_rate": 9.288888888888888e-05,
+      "loss": 0.0074,
+      "step": 210
+    },
+    {
+      "epoch": 0.8171206225680934,
+      "eval_loss": 0.011569861322641373,
+      "eval_runtime": 136.4761,
+      "eval_samples_per_second": 3.737,
+      "eval_steps_per_second": 0.938,
+      "step": 210
+    },
+    {
+      "epoch": 0.8365758754863813,
+      "grad_norm": 0.036366503685712814,
+      "learning_rate": 9.511111111111112e-05,
+      "loss": 0.007,
+      "step": 215
+    },
+    {
+      "epoch": 0.8560311284046692,
+      "grad_norm": 0.07178617268800735,
+      "learning_rate": 9.733333333333333e-05,
+      "loss": 0.0073,
+      "step": 220
+    },
+    {
+      "epoch": 0.8560311284046692,
+      "eval_loss": 0.010845971293747425,
+      "eval_runtime": 136.7794,
+      "eval_samples_per_second": 3.729,
+      "eval_steps_per_second": 0.936,
+      "step": 220
+    },
+    {
+      "epoch": 0.8754863813229572,
+      "grad_norm": 0.042044900357723236,
+      "learning_rate": 9.955555555555556e-05,
+      "loss": 0.0069,
+      "step": 225
+    },
+    {
+      "epoch": 0.8949416342412452,
+      "grad_norm": 0.18266713619232178,
+      "learning_rate": 0.00010177777777777777,
+      "loss": 0.007,
+      "step": 230
+    },
+    {
+      "epoch": 0.8949416342412452,
+      "eval_loss": 0.012411113828420639,
+      "eval_runtime": 136.7046,
+      "eval_samples_per_second": 3.731,
+      "eval_steps_per_second": 0.936,
+      "step": 230
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1799,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 10,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 30,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 9
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.3589467855146662e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-230/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfc74119a8e3587a81c422ff23316abe5d974ec5c66265926247177773d932ad
+size 6417

checkpoint-240/README.md

@@ -0,0 +1,210 @@
+---
+base_model: unsloth/medgemma-4b-it-unsloth-bnb-4bit
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:unsloth/medgemma-4b-it-unsloth-bnb-4bit
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+---
+
+# 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 0.18.0

checkpoint-240/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Gemma3ForConditionalGeneration",
+    "parent_library": "transformers.models.gemma3.modeling_gemma3",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/medgemma-4b-it-unsloth-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "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": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(?:.*?(?:vision|image|visual|patch|language|text).*?(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense).*?(?:k_proj|v_proj|q_proj|out_proj|fc1|fc2|o_proj|gate_proj|up_proj|down_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense)\\.(?:(?:k_proj|v_proj|q_proj|out_proj|fc1|fc2|o_proj|gate_proj|up_proj|down_proj)))",
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": true
+}

checkpoint-240/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d56b1cc292dcd419f5b1f27215b07e8d6e36af10bc8c7424a4061e759ccde21
+size 308107928

checkpoint-240/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

checkpoint-240/chat_template.jinja

@@ -0,0 +1,47 @@
+{{ bos_token }}
+{%- if messages[0]['role'] == 'system' -%}
+    {%- if messages[0]['content'] is string -%}
+        {%- set first_user_prefix = messages[0]['content'] + '
+
+' -%}
+    {%- else -%}
+        {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
+
+' -%}
+    {%- endif -%}
+    {%- set loop_messages = messages[1:] -%}
+{%- else -%}
+    {%- set first_user_prefix = "" -%}
+    {%- set loop_messages = messages -%}
+{%- endif -%}
+{%- for message in loop_messages -%}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
+        {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
+    {%- endif -%}
+    {%- if (message['role'] == 'assistant') -%}
+        {%- set role = "model" -%}
+    {%- else -%}
+        {%- set role = message['role'] -%}
+    {%- endif -%}
+    {{ '<start_of_turn>' + role + '
+' + (first_user_prefix if loop.first else "") }}
+    {%- if message['content'] is string -%}
+        {{ message['content'] | trim }}
+    {%- elif message['content'] is iterable -%}
+        {%- for item in message['content'] -%}
+            {%- if item['type'] == 'image' -%}
+                {{ '<start_of_image>' }}
+            {%- elif item['type'] == 'text' -%}
+                {{ item['text'] | trim }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{ raise_exception("Invalid content type") }}
+    {%- endif -%}
+    {{ '<end_of_turn>
+' }}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    {{'<start_of_turn>model
+'}}
+{%- endif -%}

checkpoint-240/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4adddf2c383ea3669208865c55123631f55746bea6fad3ec70259c416d47409e
+size 157331237

checkpoint-240/preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_pan_and_scan": null,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "Gemma3ImageProcessor",
+  "image_seq_length": 256,
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "pan_and_scan_max_num_crops": null,
+  "pan_and_scan_min_crop_size": null,
+  "pan_and_scan_min_ratio_to_activate": null,
+  "processor_class": "Gemma3Processor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 896,
+    "width": 896
+  }
+}

checkpoint-240/processor_config.json

@@ -0,0 +1,4 @@
+{
+  "image_seq_length": 256,
+  "processor_class": "Gemma3Processor"
+}

checkpoint-240/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d669fec2ccf5ca3ce9213f9c9a45b5aeec6243c9519de15665d4a5578f882b14
+size 14645

checkpoint-240/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e1e812919a6de0bf80382e9b084548427a8615758cc093ed3ff93b68383168b
+size 1465

checkpoint-240/special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<end_of_turn>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "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-240/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7666402c0617d170e6b0a985b3130c3fb0795393aa0970600994a5d9aae12351
+size 33384822

checkpoint-240/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

checkpoint-240/trainer_state.json

@@ -0,0 +1,571 @@
+{
+  "best_global_step": 220,
+  "best_metric": 0.010845971293747425,
+  "best_model_checkpoint": "/teamspace/studios/this_studio/DATN/output/medgemma_finetuned/checkpoint-220",
+  "epoch": 0.933852140077821,
+  "eval_steps": 10,
+  "global_step": 240,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.019455252918287938,
+      "grad_norm": 3.7302613258361816,
+      "learning_rate": 1.777777777777778e-06,
+      "loss": 0.8239,
+      "step": 5
+    },
+    {
+      "epoch": 0.038910505836575876,
+      "grad_norm": 3.058382034301758,
+      "learning_rate": 4e-06,
+      "loss": 0.7964,
+      "step": 10
+    },
+    {
+      "epoch": 0.038910505836575876,
+      "eval_loss": 0.7571244239807129,
+      "eval_runtime": 152.7626,
+      "eval_samples_per_second": 3.339,
+      "eval_steps_per_second": 0.838,
+      "step": 10
+    },
+    {
+      "epoch": 0.058365758754863814,
+      "grad_norm": 1.8304738998413086,
+      "learning_rate": 6.222222222222222e-06,
+      "loss": 0.6943,
+      "step": 15
+    },
+    {
+      "epoch": 0.07782101167315175,
+      "grad_norm": 1.3162634372711182,
+      "learning_rate": 8.444444444444446e-06,
+      "loss": 0.5569,
+      "step": 20
+    },
+    {
+      "epoch": 0.07782101167315175,
+      "eval_loss": 0.5357815623283386,
+      "eval_runtime": 136.4123,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 20
+    },
+    {
+      "epoch": 0.09727626459143969,
+      "grad_norm": 1.1865513324737549,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.4222,
+      "step": 25
+    },
+    {
+      "epoch": 0.11673151750972763,
+      "grad_norm": 1.2043355703353882,
+      "learning_rate": 1.2888888888888889e-05,
+      "loss": 0.2831,
+      "step": 30
+    },
+    {
+      "epoch": 0.11673151750972763,
+      "eval_loss": 0.28264933824539185,
+      "eval_runtime": 136.5183,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 30
+    },
+    {
+      "epoch": 0.13618677042801555,
+      "grad_norm": 0.9503483772277832,
+      "learning_rate": 1.511111111111111e-05,
+      "loss": 0.1405,
+      "step": 35
+    },
+    {
+      "epoch": 0.1556420233463035,
+      "grad_norm": 0.34179583191871643,
+      "learning_rate": 1.7333333333333332e-05,
+      "loss": 0.0515,
+      "step": 40
+    },
+    {
+      "epoch": 0.1556420233463035,
+      "eval_loss": 0.05298379808664322,
+      "eval_runtime": 136.391,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 40
+    },
+    {
+      "epoch": 0.17509727626459143,
+      "grad_norm": 0.24818392097949982,
+      "learning_rate": 1.9555555555555557e-05,
+      "loss": 0.0302,
+      "step": 45
+    },
+    {
+      "epoch": 0.19455252918287938,
+      "grad_norm": 0.19131316244602203,
+      "learning_rate": 2.177777777777778e-05,
+      "loss": 0.0221,
+      "step": 50
+    },
+    {
+      "epoch": 0.19455252918287938,
+      "eval_loss": 0.024967821314930916,
+      "eval_runtime": 136.5245,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 50
+    },
+    {
+      "epoch": 0.2140077821011673,
+      "grad_norm": 2.454702377319336,
+      "learning_rate": 2.4e-05,
+      "loss": 0.0169,
+      "step": 55
+    },
+    {
+      "epoch": 0.23346303501945526,
+      "grad_norm": 0.12364993244409561,
+      "learning_rate": 2.6222222222222226e-05,
+      "loss": 0.0164,
+      "step": 60
+    },
+    {
+      "epoch": 0.23346303501945526,
+      "eval_loss": 0.021661706268787384,
+      "eval_runtime": 136.8436,
+      "eval_samples_per_second": 3.727,
+      "eval_steps_per_second": 0.935,
+      "step": 60
+    },
+    {
+      "epoch": 0.2529182879377432,
+      "grad_norm": 0.14085163176059723,
+      "learning_rate": 2.8444444444444447e-05,
+      "loss": 0.0131,
+      "step": 65
+    },
+    {
+      "epoch": 0.2723735408560311,
+      "grad_norm": 0.15322668850421906,
+      "learning_rate": 3.066666666666666e-05,
+      "loss": 0.0089,
+      "step": 70
+    },
+    {
+      "epoch": 0.2723735408560311,
+      "eval_loss": 0.016315119341015816,
+      "eval_runtime": 136.2077,
+      "eval_samples_per_second": 3.744,
+      "eval_steps_per_second": 0.94,
+      "step": 70
+    },
+    {
+      "epoch": 0.2918287937743191,
+      "grad_norm": 0.08343034237623215,
+      "learning_rate": 3.288888888888889e-05,
+      "loss": 0.0076,
+      "step": 75
+    },
+    {
+      "epoch": 0.311284046692607,
+      "grad_norm": 0.11078440397977829,
+      "learning_rate": 3.511111111111111e-05,
+      "loss": 0.008,
+      "step": 80
+    },
+    {
+      "epoch": 0.311284046692607,
+      "eval_loss": 0.015718888491392136,
+      "eval_runtime": 136.7712,
+      "eval_samples_per_second": 3.729,
+      "eval_steps_per_second": 0.936,
+      "step": 80
+    },
+    {
+      "epoch": 0.33073929961089493,
+      "grad_norm": 0.08361168950796127,
+      "learning_rate": 3.733333333333334e-05,
+      "loss": 0.008,
+      "step": 85
+    },
+    {
+      "epoch": 0.35019455252918286,
+      "grad_norm": 0.06539439409971237,
+      "learning_rate": 3.9555555555555556e-05,
+      "loss": 0.0083,
+      "step": 90
+    },
+    {
+      "epoch": 0.35019455252918286,
+      "eval_loss": 0.015910081565380096,
+      "eval_runtime": 136.5187,
+      "eval_samples_per_second": 3.736,
+      "eval_steps_per_second": 0.938,
+      "step": 90
+    },
+    {
+      "epoch": 0.36964980544747084,
+      "grad_norm": 0.14973388612270355,
+      "learning_rate": 4.177777777777778e-05,
+      "loss": 0.0085,
+      "step": 95
+    },
+    {
+      "epoch": 0.38910505836575876,
+      "grad_norm": 0.08519362658262253,
+      "learning_rate": 4.4e-05,
+      "loss": 0.0077,
+      "step": 100
+    },
+    {
+      "epoch": 0.38910505836575876,
+      "eval_loss": 0.01615685597062111,
+      "eval_runtime": 136.4452,
+      "eval_samples_per_second": 3.738,
+      "eval_steps_per_second": 0.938,
+      "step": 100
+    },
+    {
+      "epoch": 0.4085603112840467,
+      "grad_norm": 0.05565109848976135,
+      "learning_rate": 4.6222222222222224e-05,
+      "loss": 0.0085,
+      "step": 105
+    },
+    {
+      "epoch": 0.4280155642023346,
+      "grad_norm": 0.07286959886550903,
+      "learning_rate": 4.844444444444445e-05,
+      "loss": 0.0082,
+      "step": 110
+    },
+    {
+      "epoch": 0.4280155642023346,
+      "eval_loss": 0.015698084607720375,
+      "eval_runtime": 136.3832,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.939,
+      "step": 110
+    },
+    {
+      "epoch": 0.4474708171206226,
+      "grad_norm": 0.13329896330833435,
+      "learning_rate": 5.066666666666667e-05,
+      "loss": 0.0085,
+      "step": 115
+    },
+    {
+      "epoch": 0.4669260700389105,
+      "grad_norm": 0.04628467932343483,
+      "learning_rate": 5.288888888888889e-05,
+      "loss": 0.0075,
+      "step": 120
+    },
+    {
+      "epoch": 0.4669260700389105,
+      "eval_loss": 0.015623296611011028,
+      "eval_runtime": 136.0422,
+      "eval_samples_per_second": 3.749,
+      "eval_steps_per_second": 0.941,
+      "step": 120
+    },
+    {
+      "epoch": 0.48638132295719844,
+      "grad_norm": 0.058520544320344925,
+      "learning_rate": 5.511111111111111e-05,
+      "loss": 0.0079,
+      "step": 125
+    },
+    {
+      "epoch": 0.5058365758754864,
+      "grad_norm": 0.06411632895469666,
+      "learning_rate": 5.7333333333333336e-05,
+      "loss": 0.0076,
+      "step": 130
+    },
+    {
+      "epoch": 0.5058365758754864,
+      "eval_loss": 0.014402530156075954,
+      "eval_runtime": 135.5623,
+      "eval_samples_per_second": 3.762,
+      "eval_steps_per_second": 0.944,
+      "step": 130
+    },
+    {
+      "epoch": 0.5252918287937743,
+      "grad_norm": 0.044081032276153564,
+      "learning_rate": 5.9555555555555554e-05,
+      "loss": 0.0072,
+      "step": 135
+    },
+    {
+      "epoch": 0.5447470817120622,
+      "grad_norm": 0.04867592826485634,
+      "learning_rate": 6.177777777777779e-05,
+      "loss": 0.0077,
+      "step": 140
+    },
+    {
+      "epoch": 0.5447470817120622,
+      "eval_loss": 0.012970623560249805,
+      "eval_runtime": 137.0137,
+      "eval_samples_per_second": 3.722,
+      "eval_steps_per_second": 0.934,
+      "step": 140
+    },
+    {
+      "epoch": 0.5642023346303502,
+      "grad_norm": 0.044633813202381134,
+      "learning_rate": 6.4e-05,
+      "loss": 0.0077,
+      "step": 145
+    },
+    {
+      "epoch": 0.5836575875486382,
+      "grad_norm": 0.052950419485569,
+      "learning_rate": 6.622222222222222e-05,
+      "loss": 0.008,
+      "step": 150
+    },
+    {
+      "epoch": 0.5836575875486382,
+      "eval_loss": 0.012441293336451054,
+      "eval_runtime": 136.133,
+      "eval_samples_per_second": 3.746,
+      "eval_steps_per_second": 0.94,
+      "step": 150
+    },
+    {
+      "epoch": 0.603112840466926,
+      "grad_norm": 0.039904553443193436,
+      "learning_rate": 6.844444444444445e-05,
+      "loss": 0.0078,
+      "step": 155
+    },
+    {
+      "epoch": 0.622568093385214,
+      "grad_norm": 0.05680263414978981,
+      "learning_rate": 7.066666666666667e-05,
+      "loss": 0.0074,
+      "step": 160
+    },
+    {
+      "epoch": 0.622568093385214,
+      "eval_loss": 0.01192025002092123,
+      "eval_runtime": 136.5582,
+      "eval_samples_per_second": 3.735,
+      "eval_steps_per_second": 0.937,
+      "step": 160
+    },
+    {
+      "epoch": 0.642023346303502,
+      "grad_norm": 0.05537933111190796,
+      "learning_rate": 7.288888888888888e-05,
+      "loss": 0.0076,
+      "step": 165
+    },
+    {
+      "epoch": 0.6614785992217899,
+      "grad_norm": 0.04935755953192711,
+      "learning_rate": 7.511111111111111e-05,
+      "loss": 0.0077,
+      "step": 170
+    },
+    {
+      "epoch": 0.6614785992217899,
+      "eval_loss": 0.012302271090447903,
+      "eval_runtime": 136.6757,
+      "eval_samples_per_second": 3.731,
+      "eval_steps_per_second": 0.937,
+      "step": 170
+    },
+    {
+      "epoch": 0.6809338521400778,
+      "grad_norm": 0.05575108528137207,
+      "learning_rate": 7.733333333333333e-05,
+      "loss": 0.0081,
+      "step": 175
+    },
+    {
+      "epoch": 0.7003891050583657,
+      "grad_norm": 0.0551481656730175,
+      "learning_rate": 7.955555555555556e-05,
+      "loss": 0.0081,
+      "step": 180
+    },
+    {
+      "epoch": 0.7003891050583657,
+      "eval_loss": 0.011542496271431446,
+      "eval_runtime": 136.4051,
+      "eval_samples_per_second": 3.739,
+      "eval_steps_per_second": 0.938,
+      "step": 180
+    },
+    {
+      "epoch": 0.7198443579766537,
+      "grad_norm": 0.04738597571849823,
+      "learning_rate": 8.177777777777778e-05,
+      "loss": 0.0076,
+      "step": 185
+    },
+    {
+      "epoch": 0.7392996108949417,
+      "grad_norm": 0.029748599976301193,
+      "learning_rate": 8.4e-05,
+      "loss": 0.0073,
+      "step": 190
+    },
+    {
+      "epoch": 0.7392996108949417,
+      "eval_loss": 0.012059729546308517,
+      "eval_runtime": 135.8659,
+      "eval_samples_per_second": 3.754,
+      "eval_steps_per_second": 0.942,
+      "step": 190
+    },
+    {
+      "epoch": 0.7587548638132295,
+      "grad_norm": 0.03995237499475479,
+      "learning_rate": 8.622222222222223e-05,
+      "loss": 0.0077,
+      "step": 195
+    },
+    {
+      "epoch": 0.7782101167315175,
+      "grad_norm": 0.02774854749441147,
+      "learning_rate": 8.844444444444445e-05,
+      "loss": 0.0075,
+      "step": 200
+    },
+    {
+      "epoch": 0.7782101167315175,
+      "eval_loss": 0.011773883365094662,
+      "eval_runtime": 136.231,
+      "eval_samples_per_second": 3.744,
+      "eval_steps_per_second": 0.94,
+      "step": 200
+    },
+    {
+      "epoch": 0.7976653696498055,
+      "grad_norm": 0.026570243760943413,
+      "learning_rate": 9.066666666666667e-05,
+      "loss": 0.0072,
+      "step": 205
+    },
+    {
+      "epoch": 0.8171206225680934,
+      "grad_norm": 0.047289494425058365,
+      "learning_rate": 9.288888888888888e-05,
+      "loss": 0.0074,
+      "step": 210
+    },
+    {
+      "epoch": 0.8171206225680934,
+      "eval_loss": 0.011569861322641373,
+      "eval_runtime": 136.4761,
+      "eval_samples_per_second": 3.737,
+      "eval_steps_per_second": 0.938,
+      "step": 210
+    },
+    {
+      "epoch": 0.8365758754863813,
+      "grad_norm": 0.036366503685712814,
+      "learning_rate": 9.511111111111112e-05,
+      "loss": 0.007,
+      "step": 215
+    },
+    {
+      "epoch": 0.8560311284046692,
+      "grad_norm": 0.07178617268800735,
+      "learning_rate": 9.733333333333333e-05,
+      "loss": 0.0073,
+      "step": 220
+    },
+    {
+      "epoch": 0.8560311284046692,
+      "eval_loss": 0.010845971293747425,
+      "eval_runtime": 136.7794,
+      "eval_samples_per_second": 3.729,
+      "eval_steps_per_second": 0.936,
+      "step": 220
+    },
+    {
+      "epoch": 0.8754863813229572,
+      "grad_norm": 0.042044900357723236,
+      "learning_rate": 9.955555555555556e-05,
+      "loss": 0.0069,
+      "step": 225
+    },
+    {
+      "epoch": 0.8949416342412452,
+      "grad_norm": 0.18266713619232178,
+      "learning_rate": 0.00010177777777777777,
+      "loss": 0.007,
+      "step": 230
+    },
+    {
+      "epoch": 0.8949416342412452,
+      "eval_loss": 0.012411113828420639,
+      "eval_runtime": 136.7046,
+      "eval_samples_per_second": 3.731,
+      "eval_steps_per_second": 0.936,
+      "step": 230
+    },
+    {
+      "epoch": 0.914396887159533,
+      "grad_norm": 0.04458538442850113,
+      "learning_rate": 0.00010400000000000001,
+      "loss": 0.0077,
+      "step": 235
+    },
+    {
+      "epoch": 0.933852140077821,
+      "grad_norm": 0.04118124023079872,
+      "learning_rate": 0.00010622222222222222,
+      "loss": 0.0071,
+      "step": 240
+    },
+    {
+      "epoch": 0.933852140077821,
+      "eval_loss": 0.016743971034884453,
+      "eval_runtime": 136.6604,
+      "eval_samples_per_second": 3.732,
+      "eval_steps_per_second": 0.937,
+      "step": 240
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1799,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 10,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 30,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 10
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.418065259671895e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-240/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfc74119a8e3587a81c422ff23316abe5d974ec5c66265926247177773d932ad
+size 6417