Upload folder using huggingface_hub

.gitattributes

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

README.md

@@ -0,0 +1,202 @@
+---
+base_model: vidore/colqwen2-base
+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 0.11.1

adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "vidore/colqwen2-base",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
+  "task_type": "FEATURE_EXTRACTION",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b29d5197e8cf1c54c0e469699b7ea5bd52564c409784822055ec7eb552e283e8
+size 74018232

added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.json

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

checkpoint-1847/README.md

@@ -0,0 +1,202 @@
+---
+base_model: ./models/colqwen2_base
+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 0.11.1

checkpoint-1847/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "./models/colqwen2_base",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
+  "task_type": "FEATURE_EXTRACTION",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-1847/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b29d5197e8cf1c54c0e469699b7ea5bd52564c409784822055ec7eb552e283e8
+size 74018232

checkpoint-1847/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e61eec2b8e41842d7c4d4f43bd570aac1993d0f04c27ce28ff440d77ef159c05
+size 148262384

checkpoint-1847/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f26235f3b8674882506cb0c3704899dcd91e2c52aa41804516417e5a3ed8be4
+size 14244

checkpoint-1847/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fb92e886726d2ce367403fa76f57f31884a180be2f7a0a97379184614ececee
+size 1064

checkpoint-1847/trainer_state.json

@@ -0,0 +1,1465 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 100,
+  "global_step": 1847,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.005414185165132647,
+      "grad_norm": 4.75,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.8236,
+      "step": 10
+    },
+    {
+      "epoch": 0.010828370330265295,
+      "grad_norm": 4.59375,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.8484,
+      "step": 20
+    },
+    {
+      "epoch": 0.016242555495397944,
+      "grad_norm": 5.21875,
+      "learning_rate": 6e-06,
+      "loss": 0.8014,
+      "step": 30
+    },
+    {
+      "epoch": 0.02165674066053059,
+      "grad_norm": 3.984375,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.779,
+      "step": 40
+    },
+    {
+      "epoch": 0.02707092582566324,
+      "grad_norm": 2.875,
+      "learning_rate": 1e-05,
+      "loss": 0.7153,
+      "step": 50
+    },
+    {
+      "epoch": 0.03248511099079589,
+      "grad_norm": 2.328125,
+      "learning_rate": 1.2e-05,
+      "loss": 0.6692,
+      "step": 60
+    },
+    {
+      "epoch": 0.03789929615592853,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.4000000000000001e-05,
+      "loss": 0.672,
+      "step": 70
+    },
+    {
+      "epoch": 0.04331348132106118,
+      "grad_norm": 1.9765625,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.6071,
+      "step": 80
+    },
+    {
+      "epoch": 0.04872766648619383,
+      "grad_norm": 1.8046875,
+      "learning_rate": 1.8e-05,
+      "loss": 0.5757,
+      "step": 90
+    },
+    {
+      "epoch": 0.05414185165132648,
+      "grad_norm": 1.6640625,
+      "learning_rate": 2e-05,
+      "loss": 0.5318,
+      "step": 100
+    },
+    {
+      "epoch": 0.05414185165132648,
+      "eval_loss": 0.4426738917827606,
+      "eval_runtime": 58.7167,
+      "eval_samples_per_second": 8.515,
+      "eval_steps_per_second": 0.545,
+      "step": 100
+    },
+    {
+      "epoch": 0.05955603681645912,
+      "grad_norm": 3.078125,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 0.5026,
+      "step": 110
+    },
+    {
+      "epoch": 0.06497022198159177,
+      "grad_norm": 3.234375,
+      "learning_rate": 2.4e-05,
+      "loss": 0.4925,
+      "step": 120
+    },
+    {
+      "epoch": 0.07038440714672442,
+      "grad_norm": 1.4296875,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 0.4557,
+      "step": 130
+    },
+    {
+      "epoch": 0.07579859231185707,
+      "grad_norm": 1.2734375,
+      "learning_rate": 2.8000000000000003e-05,
+      "loss": 0.4558,
+      "step": 140
+    },
+    {
+      "epoch": 0.08121277747698971,
+      "grad_norm": 1.234375,
+      "learning_rate": 3e-05,
+      "loss": 0.426,
+      "step": 150
+    },
+    {
+      "epoch": 0.08662696264212236,
+      "grad_norm": 1.203125,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.3799,
+      "step": 160
+    },
+    {
+      "epoch": 0.092041147807255,
+      "grad_norm": 1.3125,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.3866,
+      "step": 170
+    },
+    {
+      "epoch": 0.09745533297238766,
+      "grad_norm": 1.078125,
+      "learning_rate": 3.6e-05,
+      "loss": 0.3668,
+      "step": 180
+    },
+    {
+      "epoch": 0.10286951813752031,
+      "grad_norm": 1.3359375,
+      "learning_rate": 3.8e-05,
+      "loss": 0.3396,
+      "step": 190
+    },
+    {
+      "epoch": 0.10828370330265295,
+      "grad_norm": 1.1015625,
+      "learning_rate": 4e-05,
+      "loss": 0.3133,
+      "step": 200
+    },
+    {
+      "epoch": 0.10828370330265295,
+      "eval_loss": 0.23219549655914307,
+      "eval_runtime": 60.3361,
+      "eval_samples_per_second": 8.287,
+      "eval_steps_per_second": 0.53,
+      "step": 200
+    },
+    {
+      "epoch": 0.1136978884677856,
+      "grad_norm": 1.0390625,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2862,
+      "step": 210
+    },
+    {
+      "epoch": 0.11911207363291824,
+      "grad_norm": 1.1484375,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.2726,
+      "step": 220
+    },
+    {
+      "epoch": 0.12452625879805089,
+      "grad_norm": 0.9765625,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2823,
+      "step": 230
+    },
+    {
+      "epoch": 0.12994044396318355,
+      "grad_norm": 1.15625,
+      "learning_rate": 4.8e-05,
+      "loss": 0.2724,
+      "step": 240
+    },
+    {
+      "epoch": 0.1353546291283162,
+      "grad_norm": 1.359375,
+      "learning_rate": 5e-05,
+      "loss": 0.2563,
+      "step": 250
+    },
+    {
+      "epoch": 0.14076881429344884,
+      "grad_norm": 1.359375,
+      "learning_rate": 4.9686912961803384e-05,
+      "loss": 0.2243,
+      "step": 260
+    },
+    {
+      "epoch": 0.1461829994585815,
+      "grad_norm": 1.734375,
+      "learning_rate": 4.9373825923606765e-05,
+      "loss": 0.2297,
+      "step": 270
+    },
+    {
+      "epoch": 0.15159718462371413,
+      "grad_norm": 2.078125,
+      "learning_rate": 4.906073888541015e-05,
+      "loss": 0.207,
+      "step": 280
+    },
+    {
+      "epoch": 0.15701136978884678,
+      "grad_norm": 2.34375,
+      "learning_rate": 4.874765184721353e-05,
+      "loss": 0.2255,
+      "step": 290
+    },
+    {
+      "epoch": 0.16242555495397942,
+      "grad_norm": 1.1953125,
+      "learning_rate": 4.843456480901691e-05,
+      "loss": 0.1946,
+      "step": 300
+    },
+    {
+      "epoch": 0.16242555495397942,
+      "eval_loss": 0.1316749006509781,
+      "eval_runtime": 58.9058,
+      "eval_samples_per_second": 8.488,
+      "eval_steps_per_second": 0.543,
+      "step": 300
+    },
+    {
+      "epoch": 0.16783974011911207,
+      "grad_norm": 1.8671875,
+      "learning_rate": 4.812147777082029e-05,
+      "loss": 0.2035,
+      "step": 310
+    },
+    {
+      "epoch": 0.17325392528424471,
+      "grad_norm": 1.40625,
+      "learning_rate": 4.780839073262367e-05,
+      "loss": 0.1852,
+      "step": 320
+    },
+    {
+      "epoch": 0.17866811044937736,
+      "grad_norm": 1.375,
+      "learning_rate": 4.7495303694427054e-05,
+      "loss": 0.2054,
+      "step": 330
+    },
+    {
+      "epoch": 0.18408229561451,
+      "grad_norm": 2.046875,
+      "learning_rate": 4.7182216656230435e-05,
+      "loss": 0.1803,
+      "step": 340
+    },
+    {
+      "epoch": 0.18949648077964265,
+      "grad_norm": 1.6875,
+      "learning_rate": 4.6869129618033816e-05,
+      "loss": 0.1993,
+      "step": 350
+    },
+    {
+      "epoch": 0.19491066594477532,
+      "grad_norm": 1.3671875,
+      "learning_rate": 4.65560425798372e-05,
+      "loss": 0.1954,
+      "step": 360
+    },
+    {
+      "epoch": 0.20032485110990797,
+      "grad_norm": 1.03125,
+      "learning_rate": 4.624295554164057e-05,
+      "loss": 0.1802,
+      "step": 370
+    },
+    {
+      "epoch": 0.20573903627504062,
+      "grad_norm": 0.87890625,
+      "learning_rate": 4.5929868503443954e-05,
+      "loss": 0.1941,
+      "step": 380
+    },
+    {
+      "epoch": 0.21115322144017326,
+      "grad_norm": 1.234375,
+      "learning_rate": 4.561678146524734e-05,
+      "loss": 0.1706,
+      "step": 390
+    },
+    {
+      "epoch": 0.2165674066053059,
+      "grad_norm": 0.9765625,
+      "learning_rate": 4.5303694427050724e-05,
+      "loss": 0.1845,
+      "step": 400
+    },
+    {
+      "epoch": 0.2165674066053059,
+      "eval_loss": 0.11701580137014389,
+      "eval_runtime": 58.716,
+      "eval_samples_per_second": 8.516,
+      "eval_steps_per_second": 0.545,
+      "step": 400
+    },
+    {
+      "epoch": 0.22198159177043855,
+      "grad_norm": 1.5078125,
+      "learning_rate": 4.4990607388854105e-05,
+      "loss": 0.1386,
+      "step": 410
+    },
+    {
+      "epoch": 0.2273957769355712,
+      "grad_norm": 1.015625,
+      "learning_rate": 4.4677520350657486e-05,
+      "loss": 0.1599,
+      "step": 420
+    },
+    {
+      "epoch": 0.23280996210070384,
+      "grad_norm": 1.84375,
+      "learning_rate": 4.436443331246087e-05,
+      "loss": 0.1856,
+      "step": 430
+    },
+    {
+      "epoch": 0.2382241472658365,
+      "grad_norm": 0.8203125,
+      "learning_rate": 4.405134627426425e-05,
+      "loss": 0.1766,
+      "step": 440
+    },
+    {
+      "epoch": 0.24363833243096913,
+      "grad_norm": 0.88671875,
+      "learning_rate": 4.373825923606763e-05,
+      "loss": 0.1448,
+      "step": 450
+    },
+    {
+      "epoch": 0.24905251759610178,
+      "grad_norm": 1.4453125,
+      "learning_rate": 4.342517219787101e-05,
+      "loss": 0.1838,
+      "step": 460
+    },
+    {
+      "epoch": 0.25446670276123445,
+      "grad_norm": 0.87109375,
+      "learning_rate": 4.3112085159674393e-05,
+      "loss": 0.1698,
+      "step": 470
+    },
+    {
+      "epoch": 0.2598808879263671,
+      "grad_norm": 1.6328125,
+      "learning_rate": 4.2798998121477775e-05,
+      "loss": 0.1392,
+      "step": 480
+    },
+    {
+      "epoch": 0.26529507309149974,
+      "grad_norm": 1.1171875,
+      "learning_rate": 4.2485911083281156e-05,
+      "loss": 0.1769,
+      "step": 490
+    },
+    {
+      "epoch": 0.2707092582566324,
+      "grad_norm": 1.25,
+      "learning_rate": 4.217282404508454e-05,
+      "loss": 0.1506,
+      "step": 500
+    },
+    {
+      "epoch": 0.2707092582566324,
+      "eval_loss": 0.11454860121011734,
+      "eval_runtime": 58.3011,
+      "eval_samples_per_second": 8.576,
+      "eval_steps_per_second": 0.549,
+      "step": 500
+    },
+    {
+      "epoch": 0.27612344342176504,
+      "grad_norm": 0.6953125,
+      "learning_rate": 4.185973700688792e-05,
+      "loss": 0.1297,
+      "step": 510
+    },
+    {
+      "epoch": 0.2815376285868977,
+      "grad_norm": 1.3515625,
+      "learning_rate": 4.15466499686913e-05,
+      "loss": 0.1459,
+      "step": 520
+    },
+    {
+      "epoch": 0.2869518137520303,
+      "grad_norm": 1.3515625,
+      "learning_rate": 4.123356293049468e-05,
+      "loss": 0.1565,
+      "step": 530
+    },
+    {
+      "epoch": 0.292365998917163,
+      "grad_norm": 1.078125,
+      "learning_rate": 4.092047589229806e-05,
+      "loss": 0.1444,
+      "step": 540
+    },
+    {
+      "epoch": 0.2977801840822956,
+      "grad_norm": 1.1484375,
+      "learning_rate": 4.0607388854101445e-05,
+      "loss": 0.1581,
+      "step": 550
+    },
+    {
+      "epoch": 0.30319436924742826,
+      "grad_norm": 1.1875,
+      "learning_rate": 4.029430181590482e-05,
+      "loss": 0.1781,
+      "step": 560
+    },
+    {
+      "epoch": 0.3086085544125609,
+      "grad_norm": 0.8359375,
+      "learning_rate": 3.99812147777082e-05,
+      "loss": 0.1333,
+      "step": 570
+    },
+    {
+      "epoch": 0.31402273957769355,
+      "grad_norm": 0.82421875,
+      "learning_rate": 3.966812773951158e-05,
+      "loss": 0.1254,
+      "step": 580
+    },
+    {
+      "epoch": 0.3194369247428262,
+      "grad_norm": 1.625,
+      "learning_rate": 3.9355040701314964e-05,
+      "loss": 0.2092,
+      "step": 590
+    },
+    {
+      "epoch": 0.32485110990795885,
+      "grad_norm": 1.5234375,
+      "learning_rate": 3.9041953663118345e-05,
+      "loss": 0.1269,
+      "step": 600
+    },
+    {
+      "epoch": 0.32485110990795885,
+      "eval_loss": 0.1084875538945198,
+      "eval_runtime": 61.8367,
+      "eval_samples_per_second": 8.086,
+      "eval_steps_per_second": 0.517,
+      "step": 600
+    },
+    {
+      "epoch": 0.3302652950730915,
+      "grad_norm": 0.8203125,
+      "learning_rate": 3.8728866624921726e-05,
+      "loss": 0.1448,
+      "step": 610
+    },
+    {
+      "epoch": 0.33567948023822414,
+      "grad_norm": 0.90625,
+      "learning_rate": 3.841577958672511e-05,
+      "loss": 0.1529,
+      "step": 620
+    },
+    {
+      "epoch": 0.3410936654033568,
+      "grad_norm": 0.734375,
+      "learning_rate": 3.810269254852849e-05,
+      "loss": 0.1434,
+      "step": 630
+    },
+    {
+      "epoch": 0.34650785056848943,
+      "grad_norm": 1.1484375,
+      "learning_rate": 3.778960551033187e-05,
+      "loss": 0.1679,
+      "step": 640
+    },
+    {
+      "epoch": 0.3519220357336221,
+      "grad_norm": 1.1953125,
+      "learning_rate": 3.747651847213526e-05,
+      "loss": 0.1832,
+      "step": 650
+    },
+    {
+      "epoch": 0.3573362208987547,
+      "grad_norm": 0.96875,
+      "learning_rate": 3.716343143393864e-05,
+      "loss": 0.1634,
+      "step": 660
+    },
+    {
+      "epoch": 0.36275040606388737,
+      "grad_norm": 1.4140625,
+      "learning_rate": 3.685034439574202e-05,
+      "loss": 0.1799,
+      "step": 670
+    },
+    {
+      "epoch": 0.36816459122902,
+      "grad_norm": 0.6875,
+      "learning_rate": 3.65372573575454e-05,
+      "loss": 0.1308,
+      "step": 680
+    },
+    {
+      "epoch": 0.37357877639415266,
+      "grad_norm": 0.71875,
+      "learning_rate": 3.6224170319348784e-05,
+      "loss": 0.1282,
+      "step": 690
+    },
+    {
+      "epoch": 0.3789929615592853,
+      "grad_norm": 0.99609375,
+      "learning_rate": 3.5911083281152166e-05,
+      "loss": 0.159,
+      "step": 700
+    },
+    {
+      "epoch": 0.3789929615592853,
+      "eval_loss": 0.11033277213573456,
+      "eval_runtime": 59.5923,
+      "eval_samples_per_second": 8.39,
+      "eval_steps_per_second": 0.537,
+      "step": 700
+    },
+    {
+      "epoch": 0.38440714672441795,
+      "grad_norm": 1.9453125,
+      "learning_rate": 3.559799624295555e-05,
+      "loss": 0.1718,
+      "step": 710
+    },
+    {
+      "epoch": 0.38982133188955065,
+      "grad_norm": 1.328125,
+      "learning_rate": 3.528490920475893e-05,
+      "loss": 0.1584,
+      "step": 720
+    },
+    {
+      "epoch": 0.3952355170546833,
+      "grad_norm": 0.431640625,
+      "learning_rate": 3.497182216656231e-05,
+      "loss": 0.1556,
+      "step": 730
+    },
+    {
+      "epoch": 0.40064970221981594,
+      "grad_norm": 1.484375,
+      "learning_rate": 3.4658735128365685e-05,
+      "loss": 0.1379,
+      "step": 740
+    },
+    {
+      "epoch": 0.4060638873849486,
+      "grad_norm": 0.84375,
+      "learning_rate": 3.4345648090169066e-05,
+      "loss": 0.1542,
+      "step": 750
+    },
+    {
+      "epoch": 0.41147807255008123,
+      "grad_norm": 1.2578125,
+      "learning_rate": 3.403256105197245e-05,
+      "loss": 0.142,
+      "step": 760
+    },
+    {
+      "epoch": 0.4168922577152139,
+      "grad_norm": 0.92578125,
+      "learning_rate": 3.371947401377583e-05,
+      "loss": 0.1463,
+      "step": 770
+    },
+    {
+      "epoch": 0.4223064428803465,
+      "grad_norm": 1.1484375,
+      "learning_rate": 3.340638697557921e-05,
+      "loss": 0.1317,
+      "step": 780
+    },
+    {
+      "epoch": 0.42772062804547917,
+      "grad_norm": 1.2890625,
+      "learning_rate": 3.309329993738259e-05,
+      "loss": 0.1709,
+      "step": 790
+    },
+    {
+      "epoch": 0.4331348132106118,
+      "grad_norm": 0.9296875,
+      "learning_rate": 3.278021289918597e-05,
+      "loss": 0.1518,
+      "step": 800
+    },
+    {
+      "epoch": 0.4331348132106118,
+      "eval_loss": 0.10848626494407654,
+      "eval_runtime": 58.5847,
+      "eval_samples_per_second": 8.535,
+      "eval_steps_per_second": 0.546,
+      "step": 800
+    },
+    {
+      "epoch": 0.43854899837574446,
+      "grad_norm": 0.921875,
+      "learning_rate": 3.2467125860989355e-05,
+      "loss": 0.1298,
+      "step": 810
+    },
+    {
+      "epoch": 0.4439631835408771,
+      "grad_norm": 1.3046875,
+      "learning_rate": 3.2154038822792736e-05,
+      "loss": 0.1291,
+      "step": 820
+    },
+    {
+      "epoch": 0.44937736870600975,
+      "grad_norm": 0.70703125,
+      "learning_rate": 3.184095178459612e-05,
+      "loss": 0.1715,
+      "step": 830
+    },
+    {
+      "epoch": 0.4547915538711424,
+      "grad_norm": 1.375,
+      "learning_rate": 3.15278647463995e-05,
+      "loss": 0.1607,
+      "step": 840
+    },
+    {
+      "epoch": 0.46020573903627504,
+      "grad_norm": 0.671875,
+      "learning_rate": 3.121477770820288e-05,
+      "loss": 0.1261,
+      "step": 850
+    },
+    {
+      "epoch": 0.4656199242014077,
+      "grad_norm": 1.3984375,
+      "learning_rate": 3.090169067000626e-05,
+      "loss": 0.15,
+      "step": 860
+    },
+    {
+      "epoch": 0.47103410936654033,
+      "grad_norm": 0.91015625,
+      "learning_rate": 3.058860363180964e-05,
+      "loss": 0.1501,
+      "step": 870
+    },
+    {
+      "epoch": 0.476448294531673,
+      "grad_norm": 0.94140625,
+      "learning_rate": 3.0275516593613024e-05,
+      "loss": 0.1372,
+      "step": 880
+    },
+    {
+      "epoch": 0.4818624796968056,
+      "grad_norm": 1.0,
+      "learning_rate": 2.9962429555416406e-05,
+      "loss": 0.1725,
+      "step": 890
+    },
+    {
+      "epoch": 0.48727666486193827,
+      "grad_norm": 0.9609375,
+      "learning_rate": 2.9649342517219787e-05,
+      "loss": 0.1451,
+      "step": 900
+    },
+    {
+      "epoch": 0.48727666486193827,
+      "eval_loss": 0.1076415479183197,
+      "eval_runtime": 58.8592,
+      "eval_samples_per_second": 8.495,
+      "eval_steps_per_second": 0.544,
+      "step": 900
+    },
+    {
+      "epoch": 0.4926908500270709,
+      "grad_norm": 1.1796875,
+      "learning_rate": 2.9336255479023172e-05,
+      "loss": 0.1423,
+      "step": 910
+    },
+    {
+      "epoch": 0.49810503519220356,
+      "grad_norm": 1.109375,
+      "learning_rate": 2.9023168440826553e-05,
+      "loss": 0.1326,
+      "step": 920
+    },
+    {
+      "epoch": 0.5035192203573362,
+      "grad_norm": 0.9921875,
+      "learning_rate": 2.8710081402629935e-05,
+      "loss": 0.141,
+      "step": 930
+    },
+    {
+      "epoch": 0.5089334055224689,
+      "grad_norm": 2.015625,
+      "learning_rate": 2.8396994364433316e-05,
+      "loss": 0.1311,
+      "step": 940
+    },
+    {
+      "epoch": 0.5143475906876015,
+      "grad_norm": 1.234375,
+      "learning_rate": 2.8083907326236698e-05,
+      "loss": 0.1154,
+      "step": 950
+    },
+    {
+      "epoch": 0.5197617758527342,
+      "grad_norm": 0.78515625,
+      "learning_rate": 2.777082028804008e-05,
+      "loss": 0.1433,
+      "step": 960
+    },
+    {
+      "epoch": 0.5251759610178668,
+      "grad_norm": 1.546875,
+      "learning_rate": 2.745773324984346e-05,
+      "loss": 0.1601,
+      "step": 970
+    },
+    {
+      "epoch": 0.5305901461829995,
+      "grad_norm": 1.1796875,
+      "learning_rate": 2.7144646211646842e-05,
+      "loss": 0.1343,
+      "step": 980
+    },
+    {
+      "epoch": 0.5360043313481321,
+      "grad_norm": 0.953125,
+      "learning_rate": 2.683155917345022e-05,
+      "loss": 0.1467,
+      "step": 990
+    },
+    {
+      "epoch": 0.5414185165132648,
+      "grad_norm": 0.828125,
+      "learning_rate": 2.65184721352536e-05,
+      "loss": 0.1562,
+      "step": 1000
+    },
+    {
+      "epoch": 0.5414185165132648,
+      "eval_loss": 0.10366573929786682,
+      "eval_runtime": 60.3546,
+      "eval_samples_per_second": 8.284,
+      "eval_steps_per_second": 0.53,
+      "step": 1000
+    },
+    {
+      "epoch": 0.5468327016783974,
+      "grad_norm": 1.578125,
+      "learning_rate": 2.6205385097056983e-05,
+      "loss": 0.1299,
+      "step": 1010
+    },
+    {
+      "epoch": 0.5522468868435301,
+      "grad_norm": 1.1640625,
+      "learning_rate": 2.5892298058860364e-05,
+      "loss": 0.1386,
+      "step": 1020
+    },
+    {
+      "epoch": 0.5576610720086627,
+      "grad_norm": 0.9921875,
+      "learning_rate": 2.5579211020663746e-05,
+      "loss": 0.1212,
+      "step": 1030
+    },
+    {
+      "epoch": 0.5630752571737954,
+      "grad_norm": 0.671875,
+      "learning_rate": 2.5266123982467127e-05,
+      "loss": 0.1251,
+      "step": 1040
+    },
+    {
+      "epoch": 0.568489442338928,
+      "grad_norm": 1.0625,
+      "learning_rate": 2.495303694427051e-05,
+      "loss": 0.1275,
+      "step": 1050
+    },
+    {
+      "epoch": 0.5739036275040607,
+      "grad_norm": 0.66796875,
+      "learning_rate": 2.463994990607389e-05,
+      "loss": 0.1098,
+      "step": 1060
+    },
+    {
+      "epoch": 0.5793178126691932,
+      "grad_norm": 1.265625,
+      "learning_rate": 2.432686286787727e-05,
+      "loss": 0.1297,
+      "step": 1070
+    },
+    {
+      "epoch": 0.584731997834326,
+      "grad_norm": 1.5625,
+      "learning_rate": 2.4013775829680653e-05,
+      "loss": 0.1487,
+      "step": 1080
+    },
+    {
+      "epoch": 0.5901461829994585,
+      "grad_norm": 0.94921875,
+      "learning_rate": 2.3700688791484034e-05,
+      "loss": 0.1286,
+      "step": 1090
+    },
+    {
+      "epoch": 0.5955603681645912,
+      "grad_norm": 0.875,
+      "learning_rate": 2.3387601753287412e-05,
+      "loss": 0.1456,
+      "step": 1100
+    },
+    {
+      "epoch": 0.5955603681645912,
+      "eval_loss": 0.09888758510351181,
+      "eval_runtime": 62.4618,
+      "eval_samples_per_second": 8.005,
+      "eval_steps_per_second": 0.512,
+      "step": 1100
+    },
+    {
+      "epoch": 0.6009745533297238,
+      "grad_norm": 0.88671875,
+      "learning_rate": 2.3074514715090797e-05,
+      "loss": 0.1473,
+      "step": 1110
+    },
+    {
+      "epoch": 0.6063887384948565,
+      "grad_norm": 1.1953125,
+      "learning_rate": 2.2761427676894178e-05,
+      "loss": 0.155,
+      "step": 1120
+    },
+    {
+      "epoch": 0.6118029236599891,
+      "grad_norm": 0.83203125,
+      "learning_rate": 2.244834063869756e-05,
+      "loss": 0.1251,
+      "step": 1130
+    },
+    {
+      "epoch": 0.6172171088251218,
+      "grad_norm": 1.0078125,
+      "learning_rate": 2.213525360050094e-05,
+      "loss": 0.1495,
+      "step": 1140
+    },
+    {
+      "epoch": 0.6226312939902545,
+      "grad_norm": 2.109375,
+      "learning_rate": 2.1822166562304323e-05,
+      "loss": 0.142,
+      "step": 1150
+    },
+    {
+      "epoch": 0.6280454791553871,
+      "grad_norm": 1.0859375,
+      "learning_rate": 2.1509079524107704e-05,
+      "loss": 0.1303,
+      "step": 1160
+    },
+    {
+      "epoch": 0.6334596643205198,
+      "grad_norm": 0.9453125,
+      "learning_rate": 2.1195992485911085e-05,
+      "loss": 0.1263,
+      "step": 1170
+    },
+    {
+      "epoch": 0.6388738494856524,
+      "grad_norm": 0.84375,
+      "learning_rate": 2.0882905447714467e-05,
+      "loss": 0.1133,
+      "step": 1180
+    },
+    {
+      "epoch": 0.6442880346507851,
+      "grad_norm": 1.3125,
+      "learning_rate": 2.0569818409517845e-05,
+      "loss": 0.1411,
+      "step": 1190
+    },
+    {
+      "epoch": 0.6497022198159177,
+      "grad_norm": 0.90234375,
+      "learning_rate": 2.0256731371321226e-05,
+      "loss": 0.1571,
+      "step": 1200
+    },
+    {
+      "epoch": 0.6497022198159177,
+      "eval_loss": 0.09719575196504593,
+      "eval_runtime": 61.5385,
+      "eval_samples_per_second": 8.125,
+      "eval_steps_per_second": 0.52,
+      "step": 1200
+    },
+    {
+      "epoch": 0.6551164049810504,
+      "grad_norm": 0.6328125,
+      "learning_rate": 1.9943644333124608e-05,
+      "loss": 0.1521,
+      "step": 1210
+    },
+    {
+      "epoch": 0.660530590146183,
+      "grad_norm": 0.70703125,
+      "learning_rate": 1.963055729492799e-05,
+      "loss": 0.1128,
+      "step": 1220
+    },
+    {
+      "epoch": 0.6659447753113157,
+      "grad_norm": 1.0078125,
+      "learning_rate": 1.931747025673137e-05,
+      "loss": 0.134,
+      "step": 1230
+    },
+    {
+      "epoch": 0.6713589604764483,
+      "grad_norm": 1.0390625,
+      "learning_rate": 1.9004383218534755e-05,
+      "loss": 0.1263,
+      "step": 1240
+    },
+    {
+      "epoch": 0.676773145641581,
+      "grad_norm": 0.92578125,
+      "learning_rate": 1.8691296180338137e-05,
+      "loss": 0.1273,
+      "step": 1250
+    },
+    {
+      "epoch": 0.6821873308067136,
+      "grad_norm": 0.75,
+      "learning_rate": 1.8378209142141518e-05,
+      "loss": 0.1373,
+      "step": 1260
+    },
+    {
+      "epoch": 0.6876015159718463,
+      "grad_norm": 0.58984375,
+      "learning_rate": 1.80651221039449e-05,
+      "loss": 0.1242,
+      "step": 1270
+    },
+    {
+      "epoch": 0.6930157011369789,
+      "grad_norm": 0.40234375,
+      "learning_rate": 1.775203506574828e-05,
+      "loss": 0.1424,
+      "step": 1280
+    },
+    {
+      "epoch": 0.6984298863021116,
+      "grad_norm": 0.984375,
+      "learning_rate": 1.743894802755166e-05,
+      "loss": 0.1053,
+      "step": 1290
+    },
+    {
+      "epoch": 0.7038440714672441,
+      "grad_norm": 0.97265625,
+      "learning_rate": 1.712586098935504e-05,
+      "loss": 0.1195,
+      "step": 1300
+    },
+    {
+      "epoch": 0.7038440714672441,
+      "eval_loss": 0.09898315370082855,
+      "eval_runtime": 59.819,
+      "eval_samples_per_second": 8.359,
+      "eval_steps_per_second": 0.535,
+      "step": 1300
+    },
+    {
+      "epoch": 0.7092582566323768,
+      "grad_norm": 0.5625,
+      "learning_rate": 1.681277395115842e-05,
+      "loss": 0.1339,
+      "step": 1310
+    },
+    {
+      "epoch": 0.7146724417975094,
+      "grad_norm": 0.92578125,
+      "learning_rate": 1.6499686912961803e-05,
+      "loss": 0.155,
+      "step": 1320
+    },
+    {
+      "epoch": 0.7200866269626421,
+      "grad_norm": 1.421875,
+      "learning_rate": 1.6186599874765184e-05,
+      "loss": 0.1275,
+      "step": 1330
+    },
+    {
+      "epoch": 0.7255008121277747,
+      "grad_norm": 1.25,
+      "learning_rate": 1.5873512836568566e-05,
+      "loss": 0.1454,
+      "step": 1340
+    },
+    {
+      "epoch": 0.7309149972929074,
+      "grad_norm": 1.0078125,
+      "learning_rate": 1.5560425798371947e-05,
+      "loss": 0.1184,
+      "step": 1350
+    },
+    {
+      "epoch": 0.73632918245804,
+      "grad_norm": 0.546875,
+      "learning_rate": 1.5247338760175329e-05,
+      "loss": 0.1231,
+      "step": 1360
+    },
+    {
+      "epoch": 0.7417433676231727,
+      "grad_norm": 0.77734375,
+      "learning_rate": 1.4934251721978712e-05,
+      "loss": 0.1245,
+      "step": 1370
+    },
+    {
+      "epoch": 0.7471575527883053,
+      "grad_norm": 1.1484375,
+      "learning_rate": 1.4621164683782093e-05,
+      "loss": 0.1388,
+      "step": 1380
+    },
+    {
+      "epoch": 0.752571737953438,
+      "grad_norm": 1.234375,
+      "learning_rate": 1.4308077645585475e-05,
+      "loss": 0.1599,
+      "step": 1390
+    },
+    {
+      "epoch": 0.7579859231185706,
+      "grad_norm": 1.2265625,
+      "learning_rate": 1.3994990607388856e-05,
+      "loss": 0.1481,
+      "step": 1400
+    },
+    {
+      "epoch": 0.7579859231185706,
+      "eval_loss": 0.09811025857925415,
+      "eval_runtime": 58.4243,
+      "eval_samples_per_second": 8.558,
+      "eval_steps_per_second": 0.548,
+      "step": 1400
+    },
+    {
+      "epoch": 0.7634001082837033,
+      "grad_norm": 0.72265625,
+      "learning_rate": 1.3681903569192236e-05,
+      "loss": 0.1228,
+      "step": 1410
+    },
+    {
+      "epoch": 0.7688142934488359,
+      "grad_norm": 0.8359375,
+      "learning_rate": 1.3368816530995617e-05,
+      "loss": 0.1356,
+      "step": 1420
+    },
+    {
+      "epoch": 0.7742284786139686,
+      "grad_norm": 1.046875,
+      "learning_rate": 1.3055729492798999e-05,
+      "loss": 0.1431,
+      "step": 1430
+    },
+    {
+      "epoch": 0.7796426637791013,
+      "grad_norm": 0.9140625,
+      "learning_rate": 1.274264245460238e-05,
+      "loss": 0.1154,
+      "step": 1440
+    },
+    {
+      "epoch": 0.7850568489442339,
+      "grad_norm": 0.76953125,
+      "learning_rate": 1.2429555416405761e-05,
+      "loss": 0.1106,
+      "step": 1450
+    },
+    {
+      "epoch": 0.7904710341093666,
+      "grad_norm": 0.8046875,
+      "learning_rate": 1.2116468378209143e-05,
+      "loss": 0.1172,
+      "step": 1460
+    },
+    {
+      "epoch": 0.7958852192744992,
+      "grad_norm": 0.7578125,
+      "learning_rate": 1.1803381340012524e-05,
+      "loss": 0.1424,
+      "step": 1470
+    },
+    {
+      "epoch": 0.8012994044396319,
+      "grad_norm": 1.0,
+      "learning_rate": 1.1490294301815906e-05,
+      "loss": 0.1328,
+      "step": 1480
+    },
+    {
+      "epoch": 0.8067135896047645,
+      "grad_norm": 2.171875,
+      "learning_rate": 1.1177207263619287e-05,
+      "loss": 0.1363,
+      "step": 1490
+    },
+    {
+      "epoch": 0.8121277747698972,
+      "grad_norm": 0.88671875,
+      "learning_rate": 1.0864120225422668e-05,
+      "loss": 0.1255,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8121277747698972,
+      "eval_loss": 0.0966864824295044,
+      "eval_runtime": 61.9578,
+      "eval_samples_per_second": 8.07,
+      "eval_steps_per_second": 0.516,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8175419599350298,
+      "grad_norm": 1.1171875,
+      "learning_rate": 1.0551033187226048e-05,
+      "loss": 0.1397,
+      "step": 1510
+    },
+    {
+      "epoch": 0.8229561451001625,
+      "grad_norm": 1.421875,
+      "learning_rate": 1.023794614902943e-05,
+      "loss": 0.1289,
+      "step": 1520
+    },
+    {
+      "epoch": 0.828370330265295,
+      "grad_norm": 1.34375,
+      "learning_rate": 9.924859110832813e-06,
+      "loss": 0.1231,
+      "step": 1530
+    },
+    {
+      "epoch": 0.8337845154304278,
+      "grad_norm": 1.15625,
+      "learning_rate": 9.611772072636194e-06,
+      "loss": 0.1477,
+      "step": 1540
+    },
+    {
+      "epoch": 0.8391987005955603,
+      "grad_norm": 0.59765625,
+      "learning_rate": 9.298685034439576e-06,
+      "loss": 0.1228,
+      "step": 1550
+    },
+    {
+      "epoch": 0.844612885760693,
+      "grad_norm": 1.59375,
+      "learning_rate": 8.985597996242955e-06,
+      "loss": 0.1745,
+      "step": 1560
+    },
+    {
+      "epoch": 0.8500270709258256,
+      "grad_norm": 0.859375,
+      "learning_rate": 8.672510958046337e-06,
+      "loss": 0.1565,
+      "step": 1570
+    },
+    {
+      "epoch": 0.8554412560909583,
+      "grad_norm": 0.95703125,
+      "learning_rate": 8.359423919849718e-06,
+      "loss": 0.1503,
+      "step": 1580
+    },
+    {
+      "epoch": 0.8608554412560909,
+      "grad_norm": 1.6875,
+      "learning_rate": 8.0463368816531e-06,
+      "loss": 0.1569,
+      "step": 1590
+    },
+    {
+      "epoch": 0.8662696264212236,
+      "grad_norm": 0.96875,
+      "learning_rate": 7.733249843456483e-06,
+      "loss": 0.1408,
+      "step": 1600
+    },
+    {
+      "epoch": 0.8662696264212236,
+      "eval_loss": 0.09672338515520096,
+      "eval_runtime": 60.7387,
+      "eval_samples_per_second": 8.232,
+      "eval_steps_per_second": 0.527,
+      "step": 1600
+    },
+    {
+      "epoch": 0.8716838115863562,
+      "grad_norm": 1.078125,
+      "learning_rate": 7.420162805259863e-06,
+      "loss": 0.1313,
+      "step": 1610
+    },
+    {
+      "epoch": 0.8770979967514889,
+      "grad_norm": 0.482421875,
+      "learning_rate": 7.107075767063244e-06,
+      "loss": 0.1506,
+      "step": 1620
+    },
+    {
+      "epoch": 0.8825121819166215,
+      "grad_norm": 0.796875,
+      "learning_rate": 6.793988728866625e-06,
+      "loss": 0.1274,
+      "step": 1630
+    },
+    {
+      "epoch": 0.8879263670817542,
+      "grad_norm": 1.515625,
+      "learning_rate": 6.4809016906700065e-06,
+      "loss": 0.1471,
+      "step": 1640
+    },
+    {
+      "epoch": 0.8933405522468868,
+      "grad_norm": 0.6875,
+      "learning_rate": 6.167814652473388e-06,
+      "loss": 0.142,
+      "step": 1650
+    },
+    {
+      "epoch": 0.8987547374120195,
+      "grad_norm": 1.296875,
+      "learning_rate": 5.854727614276769e-06,
+      "loss": 0.1084,
+      "step": 1660
+    },
+    {
+      "epoch": 0.9041689225771521,
+      "grad_norm": 0.97265625,
+      "learning_rate": 5.54164057608015e-06,
+      "loss": 0.1044,
+      "step": 1670
+    },
+    {
+      "epoch": 0.9095831077422848,
+      "grad_norm": 0.96484375,
+      "learning_rate": 5.228553537883532e-06,
+      "loss": 0.147,
+      "step": 1680
+    },
+    {
+      "epoch": 0.9149972929074174,
+      "grad_norm": 0.4765625,
+      "learning_rate": 4.9154664996869136e-06,
+      "loss": 0.1254,
+      "step": 1690
+    },
+    {
+      "epoch": 0.9204114780725501,
+      "grad_norm": 0.7109375,
+      "learning_rate": 4.602379461490294e-06,
+      "loss": 0.1208,
+      "step": 1700
+    },
+    {
+      "epoch": 0.9204114780725501,
+      "eval_loss": 0.09627044945955276,
+      "eval_runtime": 61.5631,
+      "eval_samples_per_second": 8.122,
+      "eval_steps_per_second": 0.52,
+      "step": 1700
+    },
+    {
+      "epoch": 0.9258256632376828,
+      "grad_norm": 1.015625,
+      "learning_rate": 4.289292423293676e-06,
+      "loss": 0.1468,
+      "step": 1710
+    },
+    {
+      "epoch": 0.9312398484028154,
+      "grad_norm": 1.0546875,
+      "learning_rate": 3.976205385097057e-06,
+      "loss": 0.1231,
+      "step": 1720
+    },
+    {
+      "epoch": 0.9366540335679481,
+      "grad_norm": 1.4453125,
+      "learning_rate": 3.6631183469004384e-06,
+      "loss": 0.1548,
+      "step": 1730
+    },
+    {
+      "epoch": 0.9420682187330807,
+      "grad_norm": 1.125,
+      "learning_rate": 3.35003130870382e-06,
+      "loss": 0.1332,
+      "step": 1740
+    },
+    {
+      "epoch": 0.9474824038982134,
+      "grad_norm": 0.84765625,
+      "learning_rate": 3.036944270507201e-06,
+      "loss": 0.1382,
+      "step": 1750
+    },
+    {
+      "epoch": 0.952896589063346,
+      "grad_norm": 0.72265625,
+      "learning_rate": 2.7238572323105826e-06,
+      "loss": 0.1203,
+      "step": 1760
+    },
+    {
+      "epoch": 0.9583107742284787,
+      "grad_norm": 0.96875,
+      "learning_rate": 2.410770194113964e-06,
+      "loss": 0.1206,
+      "step": 1770
+    },
+    {
+      "epoch": 0.9637249593936112,
+      "grad_norm": 1.2578125,
+      "learning_rate": 2.0976831559173454e-06,
+      "loss": 0.1512,
+      "step": 1780
+    },
+    {
+      "epoch": 0.969139144558744,
+      "grad_norm": 1.21875,
+      "learning_rate": 1.7845961177207264e-06,
+      "loss": 0.1522,
+      "step": 1790
+    },
+    {
+      "epoch": 0.9745533297238765,
+      "grad_norm": 0.63671875,
+      "learning_rate": 1.4715090795241078e-06,
+      "loss": 0.1228,
+      "step": 1800
+    },
+    {
+      "epoch": 0.9745533297238765,
+      "eval_loss": 0.09588468819856644,
+      "eval_runtime": 61.1963,
+      "eval_samples_per_second": 8.17,
+      "eval_steps_per_second": 0.523,
+      "step": 1800
+    },
+    {
+      "epoch": 0.9799675148890092,
+      "grad_norm": 0.609375,
+      "learning_rate": 1.1584220413274892e-06,
+      "loss": 0.1349,
+      "step": 1810
+    },
+    {
+      "epoch": 0.9853817000541418,
+      "grad_norm": 0.84765625,
+      "learning_rate": 8.453350031308704e-07,
+      "loss": 0.1268,
+      "step": 1820
+    },
+    {
+      "epoch": 0.9907958852192745,
+      "grad_norm": 1.4609375,
+      "learning_rate": 5.322479649342517e-07,
+      "loss": 0.1416,
+      "step": 1830
+    },
+    {
+      "epoch": 0.9962100703844071,
+      "grad_norm": 1.03125,
+      "learning_rate": 2.1916092673763307e-07,
+      "loss": 0.1308,
+      "step": 1840
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1847,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1050860758570842e+18,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1847/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79f2ac024c40a27b9483434310575f77c9c0e3b6afd5dcfec7b089dbd6f9f330
+size 5176

git_hash.txt

@@ -0,0 +1 @@
+d9684ca595f0c6b1faee171bb4e45db9fb542450

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "ColQwen2Processor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

results.json

@@ -0,0 +1 @@
+{"validation_set": {"ndcg_at_1": 0.808, "ndcg_at_3": 0.86064, "ndcg_at_5": 0.87217, "ndcg_at_10": 0.87935, "ndcg_at_20": 0.8819, "ndcg_at_50": 0.8855, "ndcg_at_100": 0.88709, "map_at_1": 0.808, "map_at_3": 0.84833, "map_at_5": 0.85473, "map_at_10": 0.85773, "map_at_20": 0.85844, "map_at_50": 0.85904, "map_at_100": 0.85917, "recall_at_1": 0.808, "recall_at_3": 0.896, "recall_at_5": 0.924, "recall_at_10": 0.946, "recall_at_20": 0.956, "recall_at_50": 0.974, "recall_at_100": 0.984, "precision_at_1": 0.808, "precision_at_3": 0.29867, "precision_at_5": 0.1848, "precision_at_10": 0.0946, "precision_at_20": 0.0478, "precision_at_50": 0.01948, "precision_at_100": 0.00984, "mrr_at_1": 0.812, "mrr_at_3": 0.8503333333333334, "mrr_at_5": 0.8566333333333334, "mrr_at_10": 0.8593904761904761, "mrr_at_20": 0.8602524031524031, "mrr_at_50": 0.8609465715630416, "mrr_at_100": 0.8611005335017311, "naucs_at_1_max": 0.3515086743587615, "naucs_at_1_std": 0.03249771727400904, "naucs_at_1_diff1": 0.9474194820287203, "naucs_at_3_max": 0.5962878076950838, "naucs_at_3_std": 0.3982711326066551, "naucs_at_3_diff1": 0.9212715657258372, "naucs_at_5_max": 0.6614698510983347, "naucs_at_5_std": 0.6334095041525363, "naucs_at_5_diff1": 0.9211877733549554, "naucs_at_10_max": 0.6447418473562261, "naucs_at_10_std": 0.7033405954974616, "naucs_at_10_diff1": 0.9290209911124954, "naucs_at_20_max": 0.6058059587471368, "naucs_at_20_std": 0.6962906374671136, "naucs_at_20_diff1": 0.9307147101264778, "naucs_at_50_max": 0.6038210155857129, "naucs_at_50_std": 0.7816921640451006, "naucs_at_50_diff1": 0.9384112619406629, "naucs_at_100_max": 0.663923902894502, "naucs_at_100_std": 1.0, "naucs_at_100_diff1": 0.9325980392156906}, "syntheticDocQA_energy": {"ndcg_at_1": 0.93, "ndcg_at_3": 0.94893, "ndcg_at_5": 0.95666, "ndcg_at_10": 0.95956, "ndcg_at_20": 0.95956, "ndcg_at_50": 0.96171, "ndcg_at_100": 0.96171, "map_at_1": 0.93, "map_at_3": 0.945, "map_at_5": 0.949, "map_at_10": 0.95, "map_at_20": 0.95, "map_at_50": 0.95042, "map_at_100": 0.95042, "recall_at_1": 0.93, "recall_at_3": 0.96, "recall_at_5": 0.98, "recall_at_10": 0.99, "recall_at_20": 0.99, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.93, "precision_at_3": 0.32, "precision_at_5": 0.196, "precision_at_10": 0.099, "precision_at_20": 0.0495, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.93, "mrr_at_3": 0.9483333333333333, "mrr_at_5": 0.9503333333333334, "mrr_at_10": 0.9513333333333334, "mrr_at_20": 0.9513333333333334, "mrr_at_50": 0.9517878787878787, "mrr_at_100": 0.9517878787878787, "naucs_at_1_max": 0.5745631585967725, "naucs_at_1_std": -0.473856209150329, "naucs_at_1_diff1": 0.9626517273576113, "naucs_at_3_max": 0.7414799253034536, "naucs_at_3_std": -0.28408029878618163, "naucs_at_3_diff1": 0.9346405228758151, "naucs_at_5_max": 0.8692810457516353, "naucs_at_5_std": 0.24042950513538955, "naucs_at_5_diff1": 0.9346405228758136, "naucs_at_10_max": 0.8692810457516413, "naucs_at_10_std": 0.35807656395891135, "naucs_at_10_diff1": 0.8692810457516413, "naucs_at_20_max": 0.8692810457516413, "naucs_at_20_std": 0.35807656395891135, "naucs_at_20_diff1": 0.8692810457516413, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "syntheticDocQA_healthcare_industry": {"ndcg_at_1": 0.95, "ndcg_at_3": 0.98155, "ndcg_at_5": 0.98155, "ndcg_at_10": 0.98155, "ndcg_at_20": 0.98155, "ndcg_at_50": 0.98155, "ndcg_at_100": 0.98155, "map_at_1": 0.95, "map_at_3": 0.975, "map_at_5": 0.975, "map_at_10": 0.975, "map_at_20": 0.975, "map_at_50": 0.975, "map_at_100": 0.975, "recall_at_1": 0.95, "recall_at_3": 1.0, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.95, "precision_at_3": 0.33333, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.95, "mrr_at_3": 0.975, "mrr_at_5": 0.975, "mrr_at_10": 0.975, "mrr_at_20": 0.975, "mrr_at_50": 0.975, "mrr_at_100": 0.975, "naucs_at_1_max": 0.7777777777777775, "naucs_at_1_std": 0.1309056956115747, "naucs_at_1_diff1": 1.0, "naucs_at_3_max": 1.0, "naucs_at_3_std": 1.0, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "syntheticDocQA_artificial_intelligence_test": {"ndcg_at_1": 0.98, "ndcg_at_3": 0.98631, "ndcg_at_5": 0.99062, "ndcg_at_10": 0.99062, "ndcg_at_20": 0.99062, "ndcg_at_50": 0.99062, "ndcg_at_100": 0.99062, "map_at_1": 0.98, "map_at_3": 0.985, "map_at_5": 0.9875, "map_at_10": 0.9875, "map_at_20": 0.9875, "map_at_50": 0.9875, "map_at_100": 0.9875, "recall_at_1": 0.98, "recall_at_3": 0.99, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.98, "precision_at_3": 0.33, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.98, "mrr_at_3": 0.985, "mrr_at_5": 0.9875, "mrr_at_10": 0.9875, "mrr_at_20": 0.9875, "mrr_at_50": 0.9875, "mrr_at_100": 0.9875, "naucs_at_1_max": 0.7770774976657365, "naucs_at_1_std": -0.22035480859010095, "naucs_at_1_diff1": 0.9346405228758133, "naucs_at_3_max": 1.0, "naucs_at_3_std": 0.12278244631183229, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "syntheticDocQA_government_reports": {"ndcg_at_1": 0.86, "ndcg_at_3": 0.93309, "ndcg_at_5": 0.9374, "ndcg_at_10": 0.94055, "ndcg_at_20": 0.94055, "ndcg_at_50": 0.94055, "ndcg_at_100": 0.94055, "map_at_1": 0.86, "map_at_3": 0.91667, "map_at_5": 0.91917, "map_at_10": 0.92042, "map_at_20": 0.92042, "map_at_50": 0.92042, "map_at_100": 0.92042, "recall_at_1": 0.86, "recall_at_3": 0.98, "recall_at_5": 0.99, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.86, "precision_at_3": 0.32667, "precision_at_5": 0.198, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.87, "mrr_at_3": 0.925, "mrr_at_5": 0.925, "mrr_at_10": 0.92625, "mrr_at_20": 0.92625, "mrr_at_50": 0.92625, "mrr_at_100": 0.92625, "naucs_at_1_max": 0.6605930442011902, "naucs_at_1_std": 0.45541776361368935, "naucs_at_1_diff1": 0.880836912844672, "naucs_at_3_max": 1.0, "naucs_at_3_std": 0.5401493930905577, "naucs_at_3_diff1": 0.8692810457516296, "naucs_at_5_max": 1.0, "naucs_at_5_std": 0.7222222222222276, "naucs_at_5_diff1": 0.8692810457516413, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "infovqa_subsampled": {"ndcg_at_1": 0.85, "ndcg_at_3": 0.88928, "ndcg_at_5": 0.89918, "ndcg_at_10": 0.90366, "ndcg_at_20": 0.90816, "ndcg_at_50": 0.9101, "ndcg_at_100": 0.91073, "map_at_1": 0.85, "map_at_3": 0.88, "map_at_5": 0.8855, "map_at_10": 0.88732, "map_at_20": 0.88853, "map_at_50": 0.88883, "map_at_100": 0.88887, "recall_at_1": 0.85, "recall_at_3": 0.916, "recall_at_5": 0.94, "recall_at_10": 0.954, "recall_at_20": 0.972, "recall_at_50": 0.982, "recall_at_100": 0.986, "precision_at_1": 0.85, "precision_at_3": 0.30533, "precision_at_5": 0.188, "precision_at_10": 0.0954, "precision_at_20": 0.0486, "precision_at_50": 0.01964, "precision_at_100": 0.00986, "mrr_at_1": 0.846, "mrr_at_3": 0.8786666666666666, "mrr_at_5": 0.8841666666666667, "mrr_at_10": 0.8859690476190476, "mrr_at_20": 0.8871696210222526, "mrr_at_50": 0.887457817094473, "mrr_at_100": 0.8875065832519349, "naucs_at_1_max": 0.6576374877969413, "naucs_at_1_std": 0.09139602993817167, "naucs_at_1_diff1": 0.9287341360234302, "naucs_at_3_max": 0.5740851896314072, "naucs_at_3_std": -0.0630919034280346, "naucs_at_3_diff1": 0.8925125605797886, "naucs_at_5_max": 0.8596327419856851, "naucs_at_5_std": 0.36778711484593446, "naucs_at_5_diff1": 0.8903672580143169, "naucs_at_10_max": 0.9162708561685533, "naucs_at_10_std": 0.6028498355864041, "naucs_at_10_diff1": 0.9102829537612158, "naucs_at_20_max": 0.9354741896758731, "naucs_at_20_std": 0.6860744297719135, "naucs_at_20_diff1": 0.9428104575163434, "naucs_at_50_max": 0.9709513435003653, "naucs_at_50_std": 0.8124805477746678, "naucs_at_50_diff1": 0.9564270152505423, "naucs_at_100_max": 0.9626517273576126, "naucs_at_100_std": 0.8382686407896341, "naucs_at_100_diff1": 0.9439775910364117}, "docvqa_subsampled": {"ndcg_at_1": 0.45, "ndcg_at_3": 0.5229, "ndcg_at_5": 0.54071, "ndcg_at_10": 0.56169, "ndcg_at_20": 0.57183, "ndcg_at_50": 0.58375, "ndcg_at_100": 0.5922, "map_at_1": 0.45, "map_at_3": 0.504, "map_at_5": 0.5137, "map_at_10": 0.52214, "map_at_20": 0.52493, "map_at_50": 0.52686, "map_at_100": 0.5276, "recall_at_1": 0.45, "recall_at_3": 0.578, "recall_at_5": 0.622, "recall_at_10": 0.688, "recall_at_20": 0.728, "recall_at_50": 0.788, "recall_at_100": 0.84, "precision_at_1": 0.45, "precision_at_3": 0.19267, "precision_at_5": 0.1244, "precision_at_10": 0.0688, "precision_at_20": 0.0364, "precision_at_50": 0.01576, "precision_at_100": 0.0084, "mrr_at_1": 0.452, "mrr_at_3": 0.5063333333333333, "mrr_at_5": 0.5163333333333334, "mrr_at_10": 0.5238119047619048, "mrr_at_20": 0.5271139985401595, "mrr_at_50": 0.5290200749111619, "mrr_at_100": 0.5299254540408582, "naucs_at_1_max": 0.5212021213906897, "naucs_at_1_std": 0.6806049891966216, "naucs_at_1_diff1": 0.8806495122110918, "naucs_at_3_max": 0.47156835207348163, "naucs_at_3_std": 0.8128002501903935, "naucs_at_3_diff1": 0.8088765466775932, "naucs_at_5_max": 0.4225936642138315, "naucs_at_5_std": 0.8492139709674136, "naucs_at_5_diff1": 0.8048344413805454, "naucs_at_10_max": 0.35700736285971163, "naucs_at_10_std": 0.8817867580935254, "naucs_at_10_diff1": 0.7719340152877696, "naucs_at_20_max": 0.28175167648851784, "naucs_at_20_std": 0.888878902036797, "naucs_at_20_diff1": 0.7607623528676154, "naucs_at_50_max": 0.23570044204275323, "naucs_at_50_std": 0.9034666462989657, "naucs_at_50_diff1": 0.7529716156996781, "naucs_at_100_max": 0.3394943544428075, "naucs_at_100_std": 0.9148932253313685, "naucs_at_100_diff1": 0.7261659302896408}, "arxivqa_subsampled": {"ndcg_at_1": 0.798, "ndcg_at_3": 0.85012, "ndcg_at_5": 0.86165, "ndcg_at_10": 0.86811, "ndcg_at_20": 0.87503, "ndcg_at_50": 0.88012, "ndcg_at_100": 0.88145, "map_at_1": 0.798, "map_at_3": 0.83767, "map_at_5": 0.84407, "map_at_10": 0.84672, "map_at_20": 0.84881, "map_at_50": 0.84959, "map_at_100": 0.84972, "recall_at_1": 0.798, "recall_at_3": 0.886, "recall_at_5": 0.914, "recall_at_10": 0.934, "recall_at_20": 0.96, "recall_at_50": 0.986, "recall_at_100": 0.994, "precision_at_1": 0.798, "precision_at_3": 0.29533, "precision_at_5": 0.1828, "precision_at_10": 0.0934, "precision_at_20": 0.048, "precision_at_50": 0.01972, "precision_at_100": 0.00994, "mrr_at_1": 0.796, "mrr_at_3": 0.837, "mrr_at_5": 0.8426, "mrr_at_10": 0.8454293650793651, "mrr_at_20": 0.8476594700621016, "mrr_at_50": 0.8484996653641844, "mrr_at_100": 0.8485908734354632, "naucs_at_1_max": 0.7429305020644663, "naucs_at_1_std": 0.12389279642324286, "naucs_at_1_diff1": 0.9221054092501996, "naucs_at_3_max": 0.7559059050726685, "naucs_at_3_std": 0.08804335159072413, "naucs_at_3_diff1": 0.9016514891704258, "naucs_at_5_max": 0.7771372114737365, "naucs_at_5_std": 0.08682387683755402, "naucs_at_5_diff1": 0.8953922654333033, "naucs_at_10_max": 0.7693036810683858, "naucs_at_10_std": 0.019508813626457466, "naucs_at_10_diff1": 0.9052853464618175, "naucs_at_20_max": 0.8720821661998126, "naucs_at_20_std": 0.4474089635854335, "naucs_at_20_diff1": 0.9319561157796447, "naucs_at_50_max": 0.8856209150326733, "naucs_at_50_std": 0.2575696945444932, "naucs_at_50_diff1": 1.0, "naucs_at_100_max": 0.9564270152505304, "naucs_at_100_std": 0.9564270152505304, "naucs_at_100_diff1": 1.0}, "tabfquad_subsampled": {"ndcg_at_1": 0.83214, "ndcg_at_3": 0.87712, "ndcg_at_5": 0.89235, "ndcg_at_10": 0.89806, "ndcg_at_20": 0.90534, "ndcg_at_50": 0.90678, "ndcg_at_100": 0.90738, "map_at_1": 0.83214, "map_at_3": 0.86667, "map_at_5": 0.87542, "map_at_10": 0.87773, "map_at_20": 0.87975, "map_at_50": 0.87999, "map_at_100": 0.88005, "recall_at_1": 0.83214, "recall_at_3": 0.90714, "recall_at_5": 0.94286, "recall_at_10": 0.96071, "recall_at_20": 0.98929, "recall_at_50": 0.99643, "recall_at_100": 1.0, "precision_at_1": 0.83214, "precision_at_3": 0.30238, "precision_at_5": 0.18857, "precision_at_10": 0.09607, "precision_at_20": 0.04946, "precision_at_50": 0.01993, "precision_at_100": 0.01, "mrr_at_1": 0.8285714285714286, "mrr_at_3": 0.8660714285714286, "mrr_at_5": 0.8739285714285714, "mrr_at_10": 0.8762386621315194, "mrr_at_20": 0.8782158477426059, "mrr_at_50": 0.8784424960942542, "mrr_at_100": 0.8785030287819054, "naucs_at_1_max": 0.4932986555362122, "naucs_at_1_std": 0.22308450615600825, "naucs_at_1_diff1": 0.8669798854793715, "naucs_at_3_max": 0.5318358112475757, "naucs_at_3_std": 0.22662141779788908, "naucs_at_3_diff1": 0.8142641672053457, "naucs_at_5_max": 0.5299077964519151, "naucs_at_5_std": 0.37307422969187604, "naucs_at_5_diff1": 0.7965102707749776, "naucs_at_10_max": 0.4598506069094349, "naucs_at_10_std": 0.4182582123758572, "naucs_at_10_diff1": 0.7816823699176663, "naucs_at_20_max": 0.6640211640211727, "naucs_at_20_std": 0.7424525365701908, "naucs_at_20_diff1": 0.807812013694365, "naucs_at_50_max": 0.8692810457515607, "naucs_at_50_std": 0.8692810457515607, "naucs_at_50_diff1": 0.5541549953314449, "naucs_at_100_max": 1.0, "naucs_at_100_std": 1.0, "naucs_at_100_diff1": 1.0}, "tatdqa": {"ndcg_at_1": 0.63139, "ndcg_at_3": 0.73814, "ndcg_at_5": 0.76197, "ndcg_at_10": 0.78059, "ndcg_at_20": 0.78639, "ndcg_at_50": 0.79151, "ndcg_at_100": 0.79287, "map_at_1": 0.63139, "map_at_3": 0.71277, "map_at_5": 0.72603, "map_at_10": 0.7338, "map_at_20": 0.73539, "map_at_50": 0.73626, "map_at_100": 0.73638, "recall_at_1": 0.63139, "recall_at_3": 0.81118, "recall_at_5": 0.86891, "recall_at_10": 0.92604, "recall_at_20": 0.94889, "recall_at_50": 0.97414, "recall_at_100": 0.98256, "precision_at_1": 0.63139, "precision_at_3": 0.27039, "precision_at_5": 0.17378, "precision_at_10": 0.0926, "precision_at_20": 0.04744, "precision_at_50": 0.01948, "precision_at_100": 0.00983, "mrr_at_1": 0.6301864101022249, "mrr_at_3": 0.7114652234916817, "mrr_at_5": 0.7254459811585489, "mrr_at_10": 0.7328489343221755, "mrr_at_20": 0.7346180226268433, "mrr_at_50": 0.7354215808563426, "mrr_at_100": 0.7355602372931767, "naucs_at_1_max": 0.15602344202646737, "naucs_at_1_std": -0.23537913065678764, "naucs_at_1_diff1": 0.7995887679987748, "naucs_at_3_max": 0.19948401483418163, "naucs_at_3_std": -0.17461182695422356, "naucs_at_3_diff1": 0.697068028325409, "naucs_at_5_max": 0.2702733206987723, "naucs_at_5_std": -0.07922361501648759, "naucs_at_5_diff1": 0.6924299266004303, "naucs_at_10_max": 0.362723107586991, "naucs_at_10_std": 0.16342176544180184, "naucs_at_10_diff1": 0.6704828732981863, "naucs_at_20_max": 0.44771077518715946, "naucs_at_20_std": 0.31862795330756494, "naucs_at_20_diff1": 0.6391884425916661, "naucs_at_50_max": 0.5273988645048515, "naucs_at_50_std": 0.6199789044521115, "naucs_at_50_diff1": 0.5861085850281988, "naucs_at_100_max": 0.5130808147698965, "naucs_at_100_std": 0.5869401169922615, "naucs_at_100_diff1": 0.5890188930623028}, "shift_project": {"ndcg_at_1": 0.73, "ndcg_at_3": 0.84702, "ndcg_at_5": 0.86293, "ndcg_at_10": 0.86942, "ndcg_at_20": 0.86942, "ndcg_at_50": 0.87131, "ndcg_at_100": 0.87131, "map_at_1": 0.73, "map_at_3": 0.81833, "map_at_5": 0.82683, "map_at_10": 0.82951, "map_at_20": 0.82951, "map_at_50": 0.82978, "map_at_100": 0.82978, "recall_at_1": 0.73, "recall_at_3": 0.93, "recall_at_5": 0.97, "recall_at_10": 0.99, "recall_at_20": 0.99, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.73, "precision_at_3": 0.31, "precision_at_5": 0.194, "precision_at_10": 0.099, "precision_at_20": 0.0495, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.76, "mrr_at_3": 0.8383333333333333, "mrr_at_5": 0.8468333333333333, "mrr_at_10": 0.8497499999999999, "mrr_at_20": 0.8497499999999999, "mrr_at_50": 0.8500131578947367, "mrr_at_100": 0.8500131578947367, "naucs_at_1_max": -0.1763746092905308, "naucs_at_1_std": -0.39348129509131297, "naucs_at_1_diff1": 0.6756374216123774, "naucs_at_3_max": -0.018540749633185687, "naucs_at_3_std": -0.1494597839135643, "naucs_at_3_diff1": 0.5246098439375775, "naucs_at_5_max": -0.20572673513849474, "naucs_at_5_std": -0.4344849050731363, "naucs_at_5_diff1": 0.7152194211017747, "naucs_at_10_max": -1.7399626517273863, "naucs_at_10_std": -1.7399626517273863, "naucs_at_10_diff1": 0.8692810457516413, "naucs_at_20_max": -1.7399626517273863, "naucs_at_20_std": -1.7399626517273863, "naucs_at_20_diff1": 0.8692810457516413, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}}

special_tokens_map.json

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

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88a3a6fcb80132f76da8aa40cdc3fccd7e5d8468ef15421f5b0c2715e85217d2
+size 11420538

tokenizer_config.json

@@ -0,0 +1,144 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "ColQwen2Processor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

training_config.yml

@@ -0,0 +1,65 @@
+config:
+  (): colpali_engine.trainer.colmodel_training.ColModelTrainingConfig
+  output_dir: !path ../../../models/colqwen2-ba16 # ba4-highres
+  processor:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2Processor
+    pretrained_model_name_or_path:  "./models/colqwen2_base" # "./models/paligemma-3b-mix-448"
+    # num_image_tokens: 2048
+    # max_length: 50
+
+  model:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2
+    pretrained_model_name_or_path: "./models/colqwen2_base"
+    torch_dtype:  !ext torch.bfloat16
+    use_cache: false
+    attn_implementation: "flash_attention_2"
+#    device_map: "auto"
+#    quantization_config:
+#      (): transformers.BitsAndBytesConfig
+#      load_in_4bit: true
+#      bnb_4bit_quant_type: "nf4"
+#      bnb_4bit_compute_dtype:  "bfloat16"
+#      bnb_4bit_use_double_quant: true
+
+  dataset_loading_func: !ext colpali_engine.utils.dataset_transformation.load_train_set
+  eval_dataset_loader: !import ../data/test_data.yaml
+
+  # max_length: 50
+  run_eval: true
+  add_suffix: true
+  loss_func:
+    (): colpali_engine.loss.late_interaction_losses.ColbertPairwiseCELoss
+  tr_args:
+    (): transformers.training_args.TrainingArguments
+    output_dir: null
+    overwrite_output_dir: true
+    num_train_epochs: 1
+    per_device_train_batch_size: 16
+    # 6 x 8 gpus = 48 batch size
+    # gradient_accumulation_steps: 4
+    per_device_eval_batch_size: 4
+    eval_strategy: "steps"
+    # dataloader_num_workers: 8
+    # bf16: true
+    save_steps: 500
+    logging_steps: 10
+    eval_steps: 100
+    warmup_steps: 250
+    learning_rate: 5e-5
+    save_total_limit: 1
+    # resume_from_checkpoint: true
+    # optim: "paged_adamw_8bit"
+
+  peft_config:
+    (): peft.LoraConfig
+    r: 32
+    lora_alpha: 32
+    lora_dropout: 0.1
+    init_lora_weights: "gaussian"
+    bias: "none"
+    task_type: "FEATURE_EXTRACTION"
+    target_modules: '(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+    # target_modules: '(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+