Add files using upload-large-folder tool

best_adapter/README.md

@@ -0,0 +1,207 @@
+---
+base_model: Models/Devstral-Small-2-24B-HS-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Devstral-Small-2-24B-HS-CPT
+- lora
+- transformers
+---
+
+# 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

best_adapter/adapter_config.json

@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Devstral-Small-2-24B-HS-CPT",
+  "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": 16,
+  "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": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj",
+    "o_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/checkpoint-2500/README.md

@@ -0,0 +1,207 @@
+---
+base_model: Models/Devstral-Small-2-24B-HS-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Devstral-Small-2-24B-HS-CPT
+- lora
+- transformers
+---
+
+# 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

checkpoints/checkpoint-2500/adapter_config.json

@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Devstral-Small-2-24B-HS-CPT",
+  "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": 16,
+  "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": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj",
+    "o_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/checkpoint-3500/README.md

@@ -0,0 +1,207 @@
+---
+base_model: Models/Devstral-Small-2-24B-HS-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Devstral-Small-2-24B-HS-CPT
+- lora
+- transformers
+---
+
+# 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

checkpoints/checkpoint-4500/README.md

@@ -0,0 +1,207 @@
+---
+base_model: Models/Devstral-Small-2-24B-HS-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Devstral-Small-2-24B-HS-CPT
+- lora
+- transformers
+---
+
+# 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

checkpoints/checkpoint-4500/adapter_config.json

@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Devstral-Small-2-24B-HS-CPT",
+  "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": 16,
+  "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": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj",
+    "o_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/checkpoint-5500/README.md

@@ -0,0 +1,207 @@
+---
+base_model: Models/Devstral-Small-2-24B-HS-CPT
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Models/Devstral-Small-2-24B-HS-CPT
+- lora
+- transformers
+---
+
+# 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

checkpoints/checkpoint-5500/adapter_config.json

@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Models/Devstral-Small-2-24B-HS-CPT",
+  "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": 16,
+  "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": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj",
+    "o_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

config_resolved.yaml

@@ -0,0 +1,102 @@
+run:
+  run_dir: ./task2file/sft_devstral_24B_v2
+  seed: 42
+wandb:
+  enabled: true
+  project: sft-training
+  entity: null
+  name: null
+  tags:
+  - sft-lora
+  - 24b-Devstral
+  notes: null
+model:
+  repo_id: ./Models/Devstral-Small-2-24B-HS-CPT
+  revision: null
+  base_local_dir: base_model
+  trust_remote_code: true
+  tokenizer_use_fast: true
+  device_map: auto
+  torch_dtype: bfloat16
+  use_4bit: false
+  bnb_4bit_quant_type: nf4
+  bnb_4bit_use_double_quant: false
+  bnb_4bit_compute_dtype: bfloat16
+  attn_implementation: null
+data:
+  train_jsonl: sft_dataset.jsonl
+  eval_jsonl: null
+  eval_split_ratio: 0.1
+  instruction_field: instruction
+  input_field: input
+  output_field: output
+  format_type: custom
+  system_prompt: "You are a Hyperswitch Rust code analyzer. Identify functions/structs\
+    \ that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain\
+    \ the data flow and why each component must change:\n- Flow: [Input \u2192 Processing\
+    \ \u2192 Output with arrows]\n- For each component: \"The [ComponentName] ([path])\
+    \ must [action] because [reason]\u2014without this, [consequence]\"\n- Explain\
+    \ coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\n\
+    add::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n\
+    1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for\
+    \ nested items: `status::StructName::Type::Name`\n3. Always explain \"must change\
+    \ because\" and \"without this\"\n3. Types of components: function, struct, enum,\
+    \ impl, trait\n4. If there is extra information (e.g., enum variants), include\
+    \ that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\
+    \n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook\
+    \ system routes events via EventClass enum. Flow: webhook \u2192 EventClass \u2192\
+    \ handler \u2192 processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass)\
+    \ must add Subscriptions variant because it defines event routing\u2014without\
+    \ this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus)\
+    \ must map to EventType because it converts status to events\u2014without this,\
+    \ status changes don't trigger webhooks. These are coupled: EventClass routes\
+    \ to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\n\
+    crates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n"
+  custom_template: '##INSTRUCTION
+
+    {instruction}<|im_end|>
+
+    {input}<|im_end|>
+
+    {output}<|im_end|>'
+  max_length: 2048
+  shuffle: true
+  num_proc: 4
+peft:
+  enabled: true
+  r: 8
+  lora_alpha: 16
+  lora_dropout: 0.05
+  bias: none
+  target_modules: auto
+train:
+  num_train_epochs: 6
+  per_device_train_batch_size: 1
+  per_device_eval_batch_size: 1
+  gradient_accumulation_steps: 8
+  learning_rate: 1e-4
+  weight_decay: 0.0
+  warmup_ratio: 0.08
+  lr_scheduler_type: cosine
+  optim: adamw_torch
+  max_grad_norm: 0.8
+  gradient_checkpointing: true
+  logging_steps: 2
+  save_strategy: steps
+  save_steps: 500
+  save_total_limit: 20
+  evaluation_strategy: steps
+  eval_steps: 100
+  load_best_model_at_end: true
+  early_stopping:
+    enabled: true
+    patience: 5
+    min_delta: 0.001
+    metric: eval_loss
+    mode: min
+  resume_from_checkpoint: auto
+merge:
+  enabled: true
+  merged_dtype: float16
+  max_shard_size: 2GB
+  output_dir: ./Models/Devstral-Small-2-24B-HS-CPT-SFT_v2

eval_final.json

@@ -0,0 +1,8 @@
+{
+  "eval_loss": 0.6706293225288391,
+  "eval_runtime": 511.6513,
+  "eval_samples_per_second": 4.118,
+  "eval_steps_per_second": 4.118,
+  "epoch": 3.2067510548523206,
+  "perplexity": 1.955467553274469
+}

wandb/debug-internal.log

@@ -0,0 +1,12 @@
+{"time":"2025-12-26T18:08:08.66103332Z","level":"INFO","msg":"stream: starting","core version":"0.23.1"}
+{"time":"2025-12-26T18:08:08.82172381Z","level":"INFO","msg":"stream: created new stream","id":"ny9q48hd"}
+{"time":"2025-12-26T18:08:08.821819478Z","level":"INFO","msg":"handler: started","stream_id":"ny9q48hd"}
+{"time":"2025-12-26T18:08:08.822049155Z","level":"INFO","msg":"stream: started","id":"ny9q48hd"}
+{"time":"2025-12-26T18:08:08.822072296Z","level":"INFO","msg":"writer: started","stream_id":"ny9q48hd"}
+{"time":"2025-12-26T18:08:08.822098276Z","level":"INFO","msg":"sender: started","stream_id":"ny9q48hd"}
+{"time":"2025-12-28T04:02:04.935383596Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2025-12-28T04:02:05.045953421Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2025-12-28T04:02:05.051806259Z","level":"INFO","msg":"stream: closing","id":"ny9q48hd"}
+{"time":"2025-12-28T04:02:05.051833004Z","level":"INFO","msg":"handler: closed","stream_id":"ny9q48hd"}
+{"time":"2025-12-28T04:02:05.051917075Z","level":"INFO","msg":"sender: closed","stream_id":"ny9q48hd"}
+{"time":"2025-12-28T04:02:05.051937152Z","level":"INFO","msg":"stream: closed","id":"ny9q48hd"}

wandb/debug.log

@@ -0,0 +1,29 @@
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Current SDK version is 0.23.1
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Configure stats pid to 190322
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Loading settings from /root/.config/wandb/settings
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Loading settings from /workspace/wandb/settings
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Loading settings from environment variables
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:setup_run_log_directory():714] Logging user logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180808-ny9q48hd/logs/debug.log
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:setup_run_log_directory():715] Logging internal logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180808-ny9q48hd/logs/debug-internal.log
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:init():841] calling init triggers
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:init():846] wandb.init called with sweep_config: {}
+config: {'model': {'repo_id': './Models/Devstral-Small-2-24B-HS-CPT', 'revision': None, 'base_local_dir': 'base_model', 'trust_remote_code': True, 'tokenizer_use_fast': True, 'device_map': 'auto', 'torch_dtype': 'bfloat16', 'use_4bit': False, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': False, 'bnb_4bit_compute_dtype': 'bfloat16', 'attn_implementation': None}, 'data': {'train_jsonl': 'sft_dataset.jsonl', 'eval_jsonl': None, 'eval_split_ratio': 0.1, 'instruction_field': 'instruction', 'input_field': 'input', 'output_field': 'output', 'format_type': 'custom', 'system_prompt': 'You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain the data flow and why each component must change:\n- Flow: [Input → Processing → Output with arrows]\n- For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"\n- Explain coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\nadd::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for nested items: `status::StructName::Type::Name`\n3. Always explain "must change because" and "without this"\n3. Types of components: function, struct, enum, impl, trait\n4. If there is extra information (e.g., enum variants), include that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don\'t trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\ncrates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n', 'custom_template': '##INSTRUCTION\n{instruction}<|im_end|>\n{input}<|im_end|>\n{output}<|im_end|>', 'max_length': 2048, 'shuffle': True, 'num_proc': 4}, 'peft': {'enabled': True, 'r': 8, 'lora_alpha': 16, 'lora_dropout': 0.05, 'bias': 'none', 'target_modules': 'auto'}, 'train': {'num_train_epochs': 6, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'learning_rate': '1e-4', 'weight_decay': 0.0, 'warmup_ratio': 0.08, 'lr_scheduler_type': 'cosine', 'optim': 'adamw_torch', 'max_grad_norm': 0.8, 'gradient_checkpointing': True, 'logging_steps': 2, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'evaluation_strategy': 'steps', 'eval_steps': 100, 'load_best_model_at_end': True, 'early_stopping': {'enabled': True, 'patience': 5, 'min_delta': 0.001, 'metric': 'eval_loss', 'mode': 'min'}, 'resume_from_checkpoint': 'auto'}, 'run_dir': 'task2file/sft_devstral_24B_v2', '_wandb': {}}
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:init():889] starting backend
+2025-12-26 18:08:08,653 INFO    MainThread:190322 [wandb_init.py:init():892] sending inform_init request
+2025-12-26 18:08:08,658 INFO    MainThread:190322 [wandb_init.py:init():900] backend started and connected
+2025-12-26 18:08:08,661 INFO    MainThread:190322 [wandb_init.py:init():970] updated telemetry
+2025-12-26 18:08:08,662 INFO    MainThread:190322 [wandb_init.py:init():994] communicating run to backend with 90.0 second timeout
+2025-12-26 18:08:09,021 INFO    MainThread:190322 [wandb_init.py:init():1041] starting run threads in backend
+2025-12-26 18:08:09,134 INFO    MainThread:190322 [wandb_run.py:_console_start():2521] atexit reg
+2025-12-26 18:08:09,134 INFO    MainThread:190322 [wandb_run.py:_redirect():2369] redirect: wrap_raw
+2025-12-26 18:08:09,135 INFO    MainThread:190322 [wandb_run.py:_redirect():2438] Wrapping output streams.
+2025-12-26 18:08:09,135 INFO    MainThread:190322 [wandb_run.py:_redirect():2461] Redirects installed.
+2025-12-26 18:08:09,138 INFO    MainThread:190322 [wandb_init.py:init():1081] run started, returning control to user process
+2025-12-26 18:08:52,955 INFO    MainThread:190322 [wandb_run.py:_config_callback():1396] config_cb None None {'peft_config': {'default': {'task_type': 'CAUSAL_LM', 'peft_type': 'LORA', 'auto_mapping': None, 'peft_version': '0.18.0', 'base_model_name_or_path': 'Models/Devstral-Small-2-24B-HS-CPT', 'revision': None, 'inference_mode': False, 'r': 8, 'target_modules': ['v_proj', 'q_proj', 'o_proj', 'k_proj'], 'exclude_modules': None, 'lora_alpha': 16, 'lora_dropout': 0.05, 'fan_in_fan_out': False, 'bias': 'none', 'use_rslora': False, 'modules_to_save': None, 'init_lora_weights': True, 'layers_to_transform': None, 'layers_pattern': None, 'rank_pattern': {}, 'alpha_pattern': {}, 'megatron_config': None, 'megatron_core': 'megatron.core', 'trainable_token_indices': None, 'loftq_config': {}, 'eva_config': None, 'corda_config': None, 'use_dora': False, 'alora_invocation_tokens': None, 'use_qalora': False, 'qalora_group_size': 16, 'layer_replication': None, 'runtime_config': {'ephemeral_gpu_offload': False}, 'lora_bias': False, 'target_parameters': None, 'arrow_config': None, 'ensure_weight_tying': False}}, 'image_token_index': 10, 'projector_hidden_act': 'gelu', 'vision_feature_layer': -1, 'vision_config': {'hidden_size': 1024, 'intermediate_size': 4096, 'num_hidden_layers': 24, 'num_attention_heads': 16, 'num_channels': 3, 'patch_size': 14, 'image_size': 1540, 'attention_dropout': 0.0, 'hidden_act': 'silu', 'head_dim': 64, 'initializer_range': 0.02, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': None, 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': None, 'eos_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': '', 'model_type': 'pixtral', 'output_attentions': False}, 'text_config': {'vocab_size': 131072, 'max_position_embeddings': 393216, 'hidden_size': 5120, 'intermediate_size': 32768, 'num_hidden_layers': 40, 'num_attention_heads': 32, 'sliding_window': None, 'head_dim': 128, 'num_key_value_heads': 8, 'hidden_act': 'silu', 'initializer_range': 0.02, 'rms_norm_eps': 1e-05, 'use_cache': True, 'attention_dropout': 0.0, 'rope_parameters': {'beta_fast': 32.0, 'beta_slow': 1.0, 'factor': 48.0, 'llama_4_scaling_beta': 0.1, 'mscale': 1.0, 'mscale_all_dim': 1.0, 'original_max_position_embeddings': 8192, 'rope_theta': 100000000.0, 'rope_type': 'yarn', 'type': 'yarn'}, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': False, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': None, 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 11, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': '', 'model_type': 'ministral3', 'output_attentions': False}, 'multimodal_projector_bias': False, 'spatial_merge_size': 2, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': False, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': ['Mistral3ForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': None, 'eos_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': 'Models/Devstral-Small-2-24B-HS-CPT', 'transformers_version': '5.0.0.dev0', 'model_type': 'mistral3', 'use_cache': False, 'output_attentions': False, 'output_dir': 'task2file/sft_devstral_24B_v2/checkpoints', 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 0.0001, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 0.8, 'num_train_epochs': 6.0, 'max_steps': -1, 'lr_scheduler_type': 'cosine', 'lr_scheduler_kwargs': None, 'warmup_ratio': 0.08, 'warmup_steps': 0.08, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': None, 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 2, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'enable_jit_checkpoint': False, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'use_cpu': False, 'seed': 42, 'data_seed': None, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': -1, 'ddp_backend': None, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 100, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'run_name': None, 'disable_tqdm': False, 'remove_unused_columns': False, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'eval_loss', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'project': 'huggingface', 'trackio_space_id': 'trackio', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'hub_revision': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'auto_find_batch_size': False, 'full_determinism': False, 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_num_input_tokens_seen': 'no', 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'liger_kernel_config': None, 'eval_use_gather_object': False, 'average_tokens_across_devices': True}
+2025-12-26 18:08:52,965 INFO    MainThread:190322 [wandb_config.py:__setitem__():154] [no run ID] config set model/num_parameters = 24022764544 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7b8940b75420>>
+2025-12-26 18:08:52,965 INFO    MainThread:190322 [wandb_run.py:_config_callback():1396] config_cb model/num_parameters 24022764544 None
+2025-12-28 04:02:04,643 INFO    MainThread:190322 [wandb_run.py:_finish():2287] finishing run sirajuddin-shaik-007/sft-training/ny9q48hd
+2025-12-28 04:02:04,645 INFO    MainThread:190322 [wandb_run.py:_atexit_cleanup():2486] got exitcode: 0
+2025-12-28 04:02:04,646 INFO    MainThread:190322 [wandb_run.py:_restore():2468] restore
+2025-12-28 04:02:04,646 INFO    MainThread:190322 [wandb_run.py:_restore():2474] restore done
+2025-12-28 04:02:05,050 INFO    MainThread:190322 [wandb_run.py:_footer_sync_info():3862] logging synced files

wandb/run-20251226_180557-p7wwl5ek/files/config.yaml

@@ -0,0 +1,173 @@
+_wandb:
+    value:
+        cli_version: 0.23.1
+        e:
+            d58ijptqwmgcs1za2j0g4cusbym7bptn:
+                args:
+                    - --config
+                    - trainer-kit/SFT/config_instruct.yaml
+                codePath: trainer-kit/SFT/run_instruct.py
+                codePathLocal: trainer-kit/SFT/run_instruct.py
+                cpu_count: 12
+                cpu_count_logical: 24
+                cudaVersion: "13.0"
+                disk:
+                    /:
+                        total: "791251738624"
+                        used: "385798254592"
+                email: shaiksirajuddin9949@gmail.com
+                executable: /workspace/llm_finetuning_env/bin/python
+                gpu: NVIDIA A100-SXM4-80GB
+                gpu_count: 2
+                gpu_nvidia:
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40
+                host: a100-2gpu-shell-session-757d587799-mfdvv
+                memory:
+                    total: "359047892992"
+                os: Linux-6.12.46+-x86_64-with-glibc2.35
+                program: /workspace/trainer-kit/SFT/run_instruct.py
+                python: CPython 3.10.12
+                root: task2file/sft_devstral_24B_v2
+                startedAt: "2025-12-26T18:05:57.725585Z"
+                writerId: d58ijptqwmgcs1za2j0g4cusbym7bptn
+        m: []
+        python_version: 3.10.12
+        t:
+            "1":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "2":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "3":
+                - 15
+                - 16
+            "4": 3.10.12
+            "5": 0.23.1
+            "6": 5.0.0.dev0
+            "12": 0.23.1
+            "13": linux-x86_64
+data:
+    value:
+        custom_template: |-
+            ##INSTRUCTION
+            {instruction}<|im_end|>
+            {input}<|im_end|>
+            {output}<|im_end|>
+        eval_jsonl: null
+        eval_split_ratio: 0.1
+        format_type: custom
+        input_field: input
+        instruction_field: instruction
+        max_length: 2048
+        num_proc: 4
+        output_field: output
+        shuffle: true
+        system_prompt: |
+            You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.
+
+            ## Output Format
+
+            ##OUTPUT
+            Explain the data flow and why each component must change:
+            - Flow: [Input → Processing → Output with arrows]
+            - For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"
+            - Explain coupling between components
+
+            ##SELECT
+            modify::crates/path/to/file.rs::impl::ComponentName
+            add::crates/another/file.rs::function::AnotherComponent
+            <EOS>
+
+            ## Rules
+
+            1. Use full paths: `remove::crates/folder/file.rs::Type::Name`
+            2. Use `::` for nested items: `status::StructName::Type::Name`
+            3. Always explain "must change because" and "without this"
+            3. Types of components: function, struct, enum, impl, trait
+            4. If there is extra information (e.g., enum variants), include that too.
+            5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>
+
+            ## Example
+
+            ##TASK
+            Add webhook subscription support
+
+            ##OUTPUT
+            The webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don't trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.
+
+            ##SELECT
+            crates/common_enums/src/enums.rs::EventClass
+            crates/common_enums/src/transformers.rs::SubscriptionStatus
+            <EOS>
+        train_jsonl: ./sft_dataset.jsonl
+model:
+    value:
+        attn_implementation: null
+        base_local_dir: base_model
+        bnb_4bit_compute_dtype: bfloat16
+        bnb_4bit_quant_type: nf4
+        bnb_4bit_use_double_quant: false
+        device_map: auto
+        repo_id: ./Models/Devstral-Small-2-24B-HS-CPT
+        revision: null
+        tokenizer_use_fast: true
+        torch_dtype: bfloat16
+        trust_remote_code: true
+        use_4bit: false
+peft:
+    value:
+        bias: none
+        enabled: true
+        lora_alpha: 16
+        lora_dropout: 0.05
+        r: 8
+        target_modules: auto
+run_dir:
+    value: task2file/sft_devstral_24B_v2
+train:
+    value:
+        early_stopping:
+            enabled: true
+            metric: eval_loss
+            min_delta: 0.001
+            mode: min
+            patience: 5
+        eval_steps: 100
+        evaluation_strategy: steps
+        gradient_accumulation_steps: 8
+        gradient_checkpointing: true
+        learning_rate: "1e-4"
+        load_best_model_at_end: true
+        logging_steps: 2
+        lr_scheduler_type: cosine
+        max_grad_norm: 0.8
+        num_train_epochs: 6
+        optim: adamw_torch
+        per_device_eval_batch_size: 1
+        per_device_train_batch_size: 1
+        resume_from_checkpoint: auto
+        save_steps: 500
+        save_strategy: steps
+        save_total_limit: 20
+        warmup_ratio: 0.08
+        weight_decay: 0

wandb/run-20251226_180557-p7wwl5ek/files/output.log

@@ -0,0 +1,34 @@
+Wandb initialized: project='sft-training', name='auto-generated'
+[info] Detected Mistral3 model architecture, loading with specific class
+Traceback (most recent call last):
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 983, in <module>
+    main()
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 849, in main
+    model, tokenizer = load_base_model_and_tokenizer(cfg, base_dir)
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 579, in load_base_model_and_tokenizer
+    model = Mistral3ForConditionalGeneration.from_pretrained(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/transformers/modeling_utils.py", line 3948, in from_pretrained
+    device_map = _get_device_map(model, device_map, max_memory, hf_quantizer)
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/transformers/integrations/accelerate.py", line 281, in _get_device_map
+    inferred_max_memory = get_balanced_memory(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/transformers/integrations/accelerate.py", line 197, in get_balanced_memory
+    max_memory = get_max_memory(max_memory)
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 804, in get_max_memory
+    _ = torch.tensor([0], device=i)
+KeyboardInterrupt
+Traceback (most recent call last):
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 983, in <module>
+    main()
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 849, in main
+    model, tokenizer = load_base_model_and_tokenizer(cfg, base_dir)
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 579, in load_base_model_and_tokenizer
+    model = Mistral3ForConditionalGeneration.from_pretrained(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/transformers/modeling_utils.py", line 3948, in from_pretrained
+    device_map = _get_device_map(model, device_map, max_memory, hf_quantizer)
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/transformers/integrations/accelerate.py", line 281, in _get_device_map
+    inferred_max_memory = get_balanced_memory(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/transformers/integrations/accelerate.py", line 197, in get_balanced_memory
+    max_memory = get_max_memory(max_memory)
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 804, in get_max_memory
+    _ = torch.tensor([0], device=i)
+KeyboardInterrupt

wandb/run-20251226_180557-p7wwl5ek/files/requirements.txt

@@ -0,0 +1,104 @@
+exceptiongroup==1.3.1
+wheel==0.45.1
+python-dateutil==2.9.0.post0
+nvidia-ml-py==13.580.82
+huggingface_hub==1.2.3
+idna==3.11
+click==8.3.1
+numpy==2.2.6
+httpx==0.28.1
+tokenizers==0.22.1
+sympy==1.13.1
+yarl==1.22.0
+async-timeout==5.0.1
+datasets==4.4.2
+platformdirs==4.5.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-nvtx-cu12==12.1.105
+smmap==5.0.2
+accelerate==1.12.0
+requests==2.32.5
+aiohttp==3.13.2
+bitsandbytes==0.49.0
+nvidia-cublas-cu12==12.1.3.1
+mpmath==1.3.0
+torchaudio==2.5.1+cu121
+nvidia-cuda-runtime-cu12==12.1.105
+typing-inspection==0.4.2
+GitPython==3.1.45
+xxhash==3.6.0
+nvidia-cusolver-cu12==11.4.5.107
+pydantic_core==2.41.5
+six==1.17.0
+torchvision==0.20.1+cu121
+typing_extensions==4.15.0
+triton==3.1.0
+charset-normalizer==3.4.4
+nvitop==1.6.1
+wandb==0.23.1
+regex==2025.11.3
+pip==25.3
+nvidia-cusparse-cu12==12.1.0.106
+pytz==2025.2
+Jinja2==3.1.6
+psutil==7.2.0
+pillow==12.0.0
+packaging==25.0
+safetensors==0.7.0
+sentry-sdk==2.48.0
+gitdb==4.0.12
+httpcore==1.0.9
+setuptools==80.9.0
+nvidia-cufft-cu12==11.0.2.54
+anyio==4.12.0
+transformers==5.0.0.dev0
+pydantic==2.12.5
+fsspec==2025.10.0
+filelock==3.20.0
+PyYAML==6.0.3
+hf-xet==1.2.0
+nvidia-cudnn-cu12==9.1.0.70
+tqdm==4.67.1
+MarkupSafe==2.1.5
+attrs==25.4.0
+nvidia-cuda-nvrtc-cu12==12.1.105
+peft==0.18.0
+aiohappyeyeballs==2.6.1
+networkx==3.4.2
+nvidia-nvjitlink-cu12==12.9.86
+certifi==2025.11.12
+pyarrow==22.0.0
+dill==0.4.0
+protobuf==6.33.2
+aiosignal==1.4.0
+frozenlist==1.8.0
+urllib3==2.6.2
+propcache==0.4.1
+tzdata==2025.3
+pandas==2.3.3
+annotated-types==0.7.0
+shellingham==1.5.4
+nvidia-nccl-cu12==2.21.5
+multidict==6.7.0
+nvidia-curand-cu12==10.3.2.106
+trl==0.26.2
+torch==2.5.1+cu121
+h11==0.16.0
+multiprocess==0.70.18
+typer-slim==0.21.0
+wheel==0.45.1
+tomli==2.0.1
+autocommand==2.2.2
+jaraco.context==5.3.0
+zipp==3.19.2
+packaging==24.2
+inflect==7.3.1
+typing_extensions==4.12.2
+platformdirs==4.2.2
+jaraco.functools==4.0.1
+jaraco.collections==5.1.0
+jaraco.text==3.12.1
+backports.tarfile==1.2.0
+more-itertools==10.3.0
+importlib_metadata==8.0.0
+typeguard==4.3.0

wandb/run-20251226_180557-p7wwl5ek/files/wandb-metadata.json

@@ -0,0 +1,47 @@
+{
+  "os":  "Linux-6.12.46+-x86_64-with-glibc2.35",
+  "python":  "CPython 3.10.12",
+  "startedAt":  "2025-12-26T18:05:57.725585Z",
+  "args":  [
+    "--config",
+    "trainer-kit/SFT/config_instruct.yaml"
+  ],
+  "program":  "/workspace/trainer-kit/SFT/run_instruct.py",
+  "codePath":  "trainer-kit/SFT/run_instruct.py",
+  "codePathLocal":  "trainer-kit/SFT/run_instruct.py",
+  "email":  "shaiksirajuddin9949@gmail.com",
+  "root":  "task2file/sft_devstral_24B_v2",
+  "host":  "a100-2gpu-shell-session-757d587799-mfdvv",
+  "executable":  "/workspace/llm_finetuning_env/bin/python",
+  "cpu_count":  12,
+  "cpu_count_logical":  24,
+  "gpu":  "NVIDIA A100-SXM4-80GB",
+  "gpu_count":  2,
+  "disk":  {
+    "/":  {
+      "total":  "791251738624",
+      "used":  "385798254592"
+    }
+  },
+  "memory":  {
+    "total":  "359047892992"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba"
+    },
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40"
+    }
+  ],
+  "cudaVersion":  "13.0",
+  "writerId":  "d58ijptqwmgcs1za2j0g4cusbym7bptn"
+}

wandb/run-20251226_180557-p7wwl5ek/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_wandb":{"runtime":2},"_runtime":2}

wandb/run-20251226_180557-p7wwl5ek/logs/debug-core.log

@@ -0,0 +1,14 @@
+{"time":"2025-12-26T18:05:57.823957218Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpht47h6p7/port-189168.txt","pid":189168,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2025-12-26T18:05:57.824696455Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":189168}
+{"time":"2025-12-26T18:05:57.82469386Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-189168-189238-3343268377/socket","Net":"unix"}}
+{"time":"2025-12-26T18:05:58.003542516Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2025-12-26T18:05:58.010768581Z","level":"INFO","msg":"handleInformInit: received","streamId":"p7wwl5ek","id":"1(@)"}
+{"time":"2025-12-26T18:05:58.174029705Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"p7wwl5ek","id":"1(@)"}
+{"time":"2025-12-26T18:06:01.288089083Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2025-12-26T18:06:01.288159025Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2025-12-26T18:06:01.288200541Z","level":"INFO","msg":"server is shutting down"}
+{"time":"2025-12-26T18:06:01.288263567Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2025-12-26T18:06:01.288324658Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-189168-189238-3343268377/socket","Net":"unix"}}
+{"time":"2025-12-26T18:06:01.731707539Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2025-12-26T18:06:01.731736159Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2025-12-26T18:06:01.731744711Z","level":"INFO","msg":"server is closed"}

wandb/run-20251226_180557-p7wwl5ek/logs/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2025-12-26T18:05:58.010910802Z","level":"INFO","msg":"stream: starting","core version":"0.23.1"}
+{"time":"2025-12-26T18:05:58.173852132Z","level":"INFO","msg":"stream: created new stream","id":"p7wwl5ek"}
+{"time":"2025-12-26T18:05:58.173936115Z","level":"INFO","msg":"handler: started","stream_id":"p7wwl5ek"}
+{"time":"2025-12-26T18:05:58.174019933Z","level":"INFO","msg":"stream: started","id":"p7wwl5ek"}
+{"time":"2025-12-26T18:05:58.174038448Z","level":"INFO","msg":"writer: started","stream_id":"p7wwl5ek"}
+{"time":"2025-12-26T18:05:58.174048363Z","level":"INFO","msg":"sender: started","stream_id":"p7wwl5ek"}
+{"time":"2025-12-26T18:06:01.288165843Z","level":"INFO","msg":"stream: closing","id":"p7wwl5ek"}
+{"time":"2025-12-26T18:06:01.633870412Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2025-12-26T18:06:01.730892428Z","level":"INFO","msg":"handler: closed","stream_id":"p7wwl5ek"}
+{"time":"2025-12-26T18:06:01.730977697Z","level":"INFO","msg":"sender: closed","stream_id":"p7wwl5ek"}
+{"time":"2025-12-26T18:06:01.730985259Z","level":"INFO","msg":"stream: closed","id":"p7wwl5ek"}

wandb/run-20251226_180557-p7wwl5ek/logs/debug.log

@@ -0,0 +1,23 @@
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_setup.py:_flush():80] Current SDK version is 0.23.1
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_setup.py:_flush():80] Configure stats pid to 189168
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_setup.py:_flush():80] Loading settings from /root/.config/wandb/settings
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_setup.py:_flush():80] Loading settings from /workspace/wandb/settings
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_setup.py:_flush():80] Loading settings from environment variables
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_init.py:setup_run_log_directory():714] Logging user logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180557-p7wwl5ek/logs/debug.log
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_init.py:setup_run_log_directory():715] Logging internal logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180557-p7wwl5ek/logs/debug-internal.log
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_init.py:init():841] calling init triggers
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_init.py:init():846] wandb.init called with sweep_config: {}
+config: {'model': {'repo_id': './Models/Devstral-Small-2-24B-HS-CPT', 'revision': None, 'base_local_dir': 'base_model', 'trust_remote_code': True, 'tokenizer_use_fast': True, 'device_map': 'auto', 'torch_dtype': 'bfloat16', 'use_4bit': False, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': False, 'bnb_4bit_compute_dtype': 'bfloat16', 'attn_implementation': None}, 'data': {'train_jsonl': './sft_dataset.jsonl', 'eval_jsonl': None, 'eval_split_ratio': 0.1, 'instruction_field': 'instruction', 'input_field': 'input', 'output_field': 'output', 'format_type': 'custom', 'system_prompt': 'You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain the data flow and why each component must change:\n- Flow: [Input → Processing → Output with arrows]\n- For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"\n- Explain coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\nadd::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for nested items: `status::StructName::Type::Name`\n3. Always explain "must change because" and "without this"\n3. Types of components: function, struct, enum, impl, trait\n4. If there is extra information (e.g., enum variants), include that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don\'t trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\ncrates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n', 'custom_template': '##INSTRUCTION\n{instruction}<|im_end|>\n{input}<|im_end|>\n{output}<|im_end|>', 'max_length': 2048, 'shuffle': True, 'num_proc': 4}, 'peft': {'enabled': True, 'r': 8, 'lora_alpha': 16, 'lora_dropout': 0.05, 'bias': 'none', 'target_modules': 'auto'}, 'train': {'num_train_epochs': 6, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'learning_rate': '1e-4', 'weight_decay': 0.0, 'warmup_ratio': 0.08, 'lr_scheduler_type': 'cosine', 'optim': 'adamw_torch', 'max_grad_norm': 0.8, 'gradient_checkpointing': True, 'logging_steps': 2, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'evaluation_strategy': 'steps', 'eval_steps': 100, 'load_best_model_at_end': True, 'early_stopping': {'enabled': True, 'patience': 5, 'min_delta': 0.001, 'metric': 'eval_loss', 'mode': 'min'}, 'resume_from_checkpoint': 'auto'}, 'run_dir': 'task2file/sft_devstral_24B_v2', '_wandb': {}}
+2025-12-26 18:05:57,727 INFO    MainThread:189168 [wandb_init.py:init():889] starting backend
+2025-12-26 18:05:58,003 INFO    MainThread:189168 [wandb_init.py:init():892] sending inform_init request
+2025-12-26 18:05:58,008 INFO    MainThread:189168 [wandb_init.py:init():900] backend started and connected
+2025-12-26 18:05:58,010 INFO    MainThread:189168 [wandb_init.py:init():970] updated telemetry
+2025-12-26 18:05:58,011 INFO    MainThread:189168 [wandb_init.py:init():994] communicating run to backend with 90.0 second timeout
+2025-12-26 18:05:58,366 INFO    MainThread:189168 [wandb_init.py:init():1041] starting run threads in backend
+2025-12-26 18:05:58,481 INFO    MainThread:189168 [wandb_run.py:_console_start():2521] atexit reg
+2025-12-26 18:05:58,481 INFO    MainThread:189168 [wandb_run.py:_redirect():2369] redirect: wrap_raw
+2025-12-26 18:05:58,481 INFO    MainThread:189168 [wandb_run.py:_redirect():2438] Wrapping output streams.
+2025-12-26 18:05:58,481 INFO    MainThread:189168 [wandb_run.py:_redirect():2461] Redirects installed.
+2025-12-26 18:05:58,485 INFO    MainThread:189168 [wandb_init.py:init():1081] run started, returning control to user process
+2025-12-26 18:06:01,288 INFO    wandb-AsyncioManager-main:189168 [service_client.py:_forward_responses():80] Reached EOF.
+2025-12-26 18:06:01,288 INFO    wandb-AsyncioManager-main:189168 [mailbox.py:close():137] Closing mailbox, abandoning 1 handles.

wandb/run-20251226_180613-i1cmzyri/files/config.yaml

@@ -0,0 +1,173 @@
+_wandb:
+    value:
+        cli_version: 0.23.1
+        e:
+            qgjoaibnrh2irresqyfv8dka3f0628ti:
+                args:
+                    - --config
+                    - trainer-kit/SFT/config_instruct.yaml
+                codePath: trainer-kit/SFT/run_instruct.py
+                codePathLocal: trainer-kit/SFT/run_instruct.py
+                cpu_count: 12
+                cpu_count_logical: 24
+                cudaVersion: "13.0"
+                disk:
+                    /:
+                        total: "791251738624"
+                        used: "386025496576"
+                email: shaiksirajuddin9949@gmail.com
+                executable: /workspace/llm_finetuning_env/bin/python
+                gpu: NVIDIA A100-SXM4-80GB
+                gpu_count: 2
+                gpu_nvidia:
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40
+                host: a100-2gpu-shell-session-757d587799-mfdvv
+                memory:
+                    total: "359047892992"
+                os: Linux-6.12.46+-x86_64-with-glibc2.35
+                program: /workspace/trainer-kit/SFT/run_instruct.py
+                python: CPython 3.10.12
+                root: task2file/sft_devstral_24B_v2
+                startedAt: "2025-12-26T18:06:13.427654Z"
+                writerId: qgjoaibnrh2irresqyfv8dka3f0628ti
+        m: []
+        python_version: 3.10.12
+        t:
+            "1":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "2":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "3":
+                - 15
+                - 16
+            "4": 3.10.12
+            "5": 0.23.1
+            "6": 5.0.0.dev0
+            "12": 0.23.1
+            "13": linux-x86_64
+data:
+    value:
+        custom_template: |-
+            ##INSTRUCTION
+            {instruction}<|im_end|>
+            {input}<|im_end|>
+            {output}<|im_end|>
+        eval_jsonl: null
+        eval_split_ratio: 0.1
+        format_type: custom
+        input_field: input
+        instruction_field: instruction
+        max_length: 2048
+        num_proc: 4
+        output_field: output
+        shuffle: true
+        system_prompt: |
+            You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.
+
+            ## Output Format
+
+            ##OUTPUT
+            Explain the data flow and why each component must change:
+            - Flow: [Input → Processing → Output with arrows]
+            - For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"
+            - Explain coupling between components
+
+            ##SELECT
+            modify::crates/path/to/file.rs::impl::ComponentName
+            add::crates/another/file.rs::function::AnotherComponent
+            <EOS>
+
+            ## Rules
+
+            1. Use full paths: `remove::crates/folder/file.rs::Type::Name`
+            2. Use `::` for nested items: `status::StructName::Type::Name`
+            3. Always explain "must change because" and "without this"
+            3. Types of components: function, struct, enum, impl, trait
+            4. If there is extra information (e.g., enum variants), include that too.
+            5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>
+
+            ## Example
+
+            ##TASK
+            Add webhook subscription support
+
+            ##OUTPUT
+            The webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don't trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.
+
+            ##SELECT
+            crates/common_enums/src/enums.rs::EventClass
+            crates/common_enums/src/transformers.rs::SubscriptionStatus
+            <EOS>
+        train_jsonl: ./sft_dataset.jsonl
+model:
+    value:
+        attn_implementation: null
+        base_local_dir: base_model
+        bnb_4bit_compute_dtype: bfloat16
+        bnb_4bit_quant_type: nf4
+        bnb_4bit_use_double_quant: false
+        device_map: auto
+        repo_id: ./Models/Devstral-Small-2-24B-HS-CPT
+        revision: null
+        tokenizer_use_fast: true
+        torch_dtype: bfloat16
+        trust_remote_code: true
+        use_4bit: false
+peft:
+    value:
+        bias: none
+        enabled: true
+        lora_alpha: 16
+        lora_dropout: 0.05
+        r: 8
+        target_modules: auto
+run_dir:
+    value: task2file/sft_devstral_24B_v2
+train:
+    value:
+        early_stopping:
+            enabled: true
+            metric: eval_loss
+            min_delta: 0.001
+            mode: min
+            patience: 5
+        eval_steps: 100
+        evaluation_strategy: steps
+        gradient_accumulation_steps: 8
+        gradient_checkpointing: true
+        learning_rate: "1e-4"
+        load_best_model_at_end: true
+        logging_steps: 2
+        lr_scheduler_type: cosine
+        max_grad_norm: 0.8
+        num_train_epochs: 6
+        optim: adamw_torch
+        per_device_eval_batch_size: 1
+        per_device_train_batch_size: 1
+        resume_from_checkpoint: auto
+        save_steps: 500
+        save_strategy: steps
+        save_total_limit: 20
+        warmup_ratio: 0.08
+        weight_decay: 0

wandb/run-20251226_180613-i1cmzyri/files/output.log

@@ -0,0 +1,48 @@
+Wandb initialized: project='sft-training', name='auto-generated'
+[info] Detected Mistral3 model architecture, loading with specific class
+Loading weights: 100%|█| 585/585 [00:12<00:00, 46.92it/s, Materializing param=model.vision_tower.transfor
+[info] Ensuring all parameters are materialized...
+Traceback (most recent call last):
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 983, in <module>
+    main()
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 852, in main
+    train_ds, eval_ds = build_datasets(cfg, tokenizer)
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 414, in build_datasets
+    ds = load_dataset("json", data_files={"train": train_path})
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 1492, in load_dataset
+    builder_instance = load_dataset_builder(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 1137, in load_dataset_builder
+    dataset_module = dataset_module_factory(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 913, in dataset_module_factory
+    ).get_module()
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 527, in get_module
+    data_files = DataFilesDict.from_patterns(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 708, in from_patterns
+    else DataFilesList.from_patterns(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 601, in from_patterns
+    resolve_pattern(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 390, in resolve_pattern
+    raise FileNotFoundError(error_msg)
+FileNotFoundError: Unable to find '/workspace/./sft_dataset.jsonl'
+Traceback (most recent call last):
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 983, in <module>
+    main()
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 852, in main
+    train_ds, eval_ds = build_datasets(cfg, tokenizer)
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 414, in build_datasets
+    ds = load_dataset("json", data_files={"train": train_path})
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 1492, in load_dataset
+    builder_instance = load_dataset_builder(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 1137, in load_dataset_builder
+    dataset_module = dataset_module_factory(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 913, in dataset_module_factory
+    ).get_module()
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 527, in get_module
+    data_files = DataFilesDict.from_patterns(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 708, in from_patterns
+    else DataFilesList.from_patterns(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 601, in from_patterns
+    resolve_pattern(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 390, in resolve_pattern
+    raise FileNotFoundError(error_msg)
+FileNotFoundError: Unable to find '/workspace/./sft_dataset.jsonl'

wandb/run-20251226_180613-i1cmzyri/files/requirements.txt

@@ -0,0 +1,104 @@
+exceptiongroup==1.3.1
+wheel==0.45.1
+python-dateutil==2.9.0.post0
+nvidia-ml-py==13.580.82
+huggingface_hub==1.2.3
+idna==3.11
+click==8.3.1
+numpy==2.2.6
+httpx==0.28.1
+tokenizers==0.22.1
+sympy==1.13.1
+yarl==1.22.0
+async-timeout==5.0.1
+datasets==4.4.2
+platformdirs==4.5.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-nvtx-cu12==12.1.105
+smmap==5.0.2
+accelerate==1.12.0
+requests==2.32.5
+aiohttp==3.13.2
+bitsandbytes==0.49.0
+nvidia-cublas-cu12==12.1.3.1
+mpmath==1.3.0
+torchaudio==2.5.1+cu121
+nvidia-cuda-runtime-cu12==12.1.105
+typing-inspection==0.4.2
+GitPython==3.1.45
+xxhash==3.6.0
+nvidia-cusolver-cu12==11.4.5.107
+pydantic_core==2.41.5
+six==1.17.0
+torchvision==0.20.1+cu121
+typing_extensions==4.15.0
+triton==3.1.0
+charset-normalizer==3.4.4
+nvitop==1.6.1
+wandb==0.23.1
+regex==2025.11.3
+pip==25.3
+nvidia-cusparse-cu12==12.1.0.106
+pytz==2025.2
+Jinja2==3.1.6
+psutil==7.2.0
+pillow==12.0.0
+packaging==25.0
+safetensors==0.7.0
+sentry-sdk==2.48.0
+gitdb==4.0.12
+httpcore==1.0.9
+setuptools==80.9.0
+nvidia-cufft-cu12==11.0.2.54
+anyio==4.12.0
+transformers==5.0.0.dev0
+pydantic==2.12.5
+fsspec==2025.10.0
+filelock==3.20.0
+PyYAML==6.0.3
+hf-xet==1.2.0
+nvidia-cudnn-cu12==9.1.0.70
+tqdm==4.67.1
+MarkupSafe==2.1.5
+attrs==25.4.0
+nvidia-cuda-nvrtc-cu12==12.1.105
+peft==0.18.0
+aiohappyeyeballs==2.6.1
+networkx==3.4.2
+nvidia-nvjitlink-cu12==12.9.86
+certifi==2025.11.12
+pyarrow==22.0.0
+dill==0.4.0
+protobuf==6.33.2
+aiosignal==1.4.0
+frozenlist==1.8.0
+urllib3==2.6.2
+propcache==0.4.1
+tzdata==2025.3
+pandas==2.3.3
+annotated-types==0.7.0
+shellingham==1.5.4
+nvidia-nccl-cu12==2.21.5
+multidict==6.7.0
+nvidia-curand-cu12==10.3.2.106
+trl==0.26.2
+torch==2.5.1+cu121
+h11==0.16.0
+multiprocess==0.70.18
+typer-slim==0.21.0
+wheel==0.45.1
+tomli==2.0.1
+autocommand==2.2.2
+jaraco.context==5.3.0
+zipp==3.19.2
+packaging==24.2
+inflect==7.3.1
+typing_extensions==4.12.2
+platformdirs==4.2.2
+jaraco.functools==4.0.1
+jaraco.collections==5.1.0
+jaraco.text==3.12.1
+backports.tarfile==1.2.0
+more-itertools==10.3.0
+importlib_metadata==8.0.0
+typeguard==4.3.0

wandb/run-20251226_180613-i1cmzyri/files/wandb-metadata.json

@@ -0,0 +1,47 @@
+{
+  "os":  "Linux-6.12.46+-x86_64-with-glibc2.35",
+  "python":  "CPython 3.10.12",
+  "startedAt":  "2025-12-26T18:06:13.427654Z",
+  "args":  [
+    "--config",
+    "trainer-kit/SFT/config_instruct.yaml"
+  ],
+  "program":  "/workspace/trainer-kit/SFT/run_instruct.py",
+  "codePath":  "trainer-kit/SFT/run_instruct.py",
+  "codePathLocal":  "trainer-kit/SFT/run_instruct.py",
+  "email":  "shaiksirajuddin9949@gmail.com",
+  "root":  "task2file/sft_devstral_24B_v2",
+  "host":  "a100-2gpu-shell-session-757d587799-mfdvv",
+  "executable":  "/workspace/llm_finetuning_env/bin/python",
+  "cpu_count":  12,
+  "cpu_count_logical":  24,
+  "gpu":  "NVIDIA A100-SXM4-80GB",
+  "gpu_count":  2,
+  "disk":  {
+    "/":  {
+      "total":  "791251738624",
+      "used":  "386025496576"
+    }
+  },
+  "memory":  {
+    "total":  "359047892992"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba"
+    },
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40"
+    }
+  ],
+  "cudaVersion":  "13.0",
+  "writerId":  "qgjoaibnrh2irresqyfv8dka3f0628ti"
+}

wandb/run-20251226_180613-i1cmzyri/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_wandb":{"runtime":19},"_runtime":19}

wandb/run-20251226_180613-i1cmzyri/logs/debug-core.log

@@ -0,0 +1,14 @@
+{"time":"2025-12-26T18:06:13.512689006Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpaql7q9l6/port-189374.txt","pid":189374,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2025-12-26T18:06:13.513337738Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":189374}
+{"time":"2025-12-26T18:06:13.513339589Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-189374-189434-2072401818/socket","Net":"unix"}}
+{"time":"2025-12-26T18:06:13.697155944Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2025-12-26T18:06:13.703520039Z","level":"INFO","msg":"handleInformInit: received","streamId":"i1cmzyri","id":"1(@)"}
+{"time":"2025-12-26T18:06:13.861987964Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"i1cmzyri","id":"1(@)"}
+{"time":"2025-12-26T18:06:33.265409361Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2025-12-26T18:06:33.265466167Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2025-12-26T18:06:33.265521907Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2025-12-26T18:06:33.265530544Z","level":"INFO","msg":"server is shutting down"}
+{"time":"2025-12-26T18:06:33.265687334Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-189374-189434-2072401818/socket","Net":"unix"}}
+{"time":"2025-12-26T18:06:33.573303185Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2025-12-26T18:06:33.573347477Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2025-12-26T18:06:33.573360632Z","level":"INFO","msg":"server is closed"}

wandb/run-20251226_180613-i1cmzyri/logs/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2025-12-26T18:06:13.703642155Z","level":"INFO","msg":"stream: starting","core version":"0.23.1"}
+{"time":"2025-12-26T18:06:13.861777831Z","level":"INFO","msg":"stream: created new stream","id":"i1cmzyri"}
+{"time":"2025-12-26T18:06:13.861855775Z","level":"INFO","msg":"handler: started","stream_id":"i1cmzyri"}
+{"time":"2025-12-26T18:06:13.861978087Z","level":"INFO","msg":"stream: started","id":"i1cmzyri"}
+{"time":"2025-12-26T18:06:13.862005472Z","level":"INFO","msg":"writer: started","stream_id":"i1cmzyri"}
+{"time":"2025-12-26T18:06:13.862018215Z","level":"INFO","msg":"sender: started","stream_id":"i1cmzyri"}
+{"time":"2025-12-26T18:06:33.265479861Z","level":"INFO","msg":"stream: closing","id":"i1cmzyri"}
+{"time":"2025-12-26T18:06:33.464312202Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2025-12-26T18:06:33.550123155Z","level":"INFO","msg":"handler: closed","stream_id":"i1cmzyri"}
+{"time":"2025-12-26T18:06:33.550255641Z","level":"INFO","msg":"sender: closed","stream_id":"i1cmzyri"}
+{"time":"2025-12-26T18:06:33.550271731Z","level":"INFO","msg":"stream: closed","id":"i1cmzyri"}

wandb/run-20251226_180613-i1cmzyri/logs/debug.log

@@ -0,0 +1,23 @@
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_setup.py:_flush():80] Current SDK version is 0.23.1
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_setup.py:_flush():80] Configure stats pid to 189374
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_setup.py:_flush():80] Loading settings from /root/.config/wandb/settings
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_setup.py:_flush():80] Loading settings from /workspace/wandb/settings
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_setup.py:_flush():80] Loading settings from environment variables
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_init.py:setup_run_log_directory():714] Logging user logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180613-i1cmzyri/logs/debug.log
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_init.py:setup_run_log_directory():715] Logging internal logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180613-i1cmzyri/logs/debug-internal.log
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_init.py:init():841] calling init triggers
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_init.py:init():846] wandb.init called with sweep_config: {}
+config: {'model': {'repo_id': './Models/Devstral-Small-2-24B-HS-CPT', 'revision': None, 'base_local_dir': 'base_model', 'trust_remote_code': True, 'tokenizer_use_fast': True, 'device_map': 'auto', 'torch_dtype': 'bfloat16', 'use_4bit': False, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': False, 'bnb_4bit_compute_dtype': 'bfloat16', 'attn_implementation': None}, 'data': {'train_jsonl': './sft_dataset.jsonl', 'eval_jsonl': None, 'eval_split_ratio': 0.1, 'instruction_field': 'instruction', 'input_field': 'input', 'output_field': 'output', 'format_type': 'custom', 'system_prompt': 'You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain the data flow and why each component must change:\n- Flow: [Input → Processing → Output with arrows]\n- For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"\n- Explain coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\nadd::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for nested items: `status::StructName::Type::Name`\n3. Always explain "must change because" and "without this"\n3. Types of components: function, struct, enum, impl, trait\n4. If there is extra information (e.g., enum variants), include that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don\'t trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\ncrates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n', 'custom_template': '##INSTRUCTION\n{instruction}<|im_end|>\n{input}<|im_end|>\n{output}<|im_end|>', 'max_length': 2048, 'shuffle': True, 'num_proc': 4}, 'peft': {'enabled': True, 'r': 8, 'lora_alpha': 16, 'lora_dropout': 0.05, 'bias': 'none', 'target_modules': 'auto'}, 'train': {'num_train_epochs': 6, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'learning_rate': '1e-4', 'weight_decay': 0.0, 'warmup_ratio': 0.08, 'lr_scheduler_type': 'cosine', 'optim': 'adamw_torch', 'max_grad_norm': 0.8, 'gradient_checkpointing': True, 'logging_steps': 2, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'evaluation_strategy': 'steps', 'eval_steps': 100, 'load_best_model_at_end': True, 'early_stopping': {'enabled': True, 'patience': 5, 'min_delta': 0.001, 'metric': 'eval_loss', 'mode': 'min'}, 'resume_from_checkpoint': 'auto'}, 'run_dir': 'task2file/sft_devstral_24B_v2', '_wandb': {}}
+2025-12-26 18:06:13,429 INFO    MainThread:189374 [wandb_init.py:init():889] starting backend
+2025-12-26 18:06:13,697 INFO    MainThread:189374 [wandb_init.py:init():892] sending inform_init request
+2025-12-26 18:06:13,701 INFO    MainThread:189374 [wandb_init.py:init():900] backend started and connected
+2025-12-26 18:06:13,703 INFO    MainThread:189374 [wandb_init.py:init():970] updated telemetry
+2025-12-26 18:06:13,704 INFO    MainThread:189374 [wandb_init.py:init():994] communicating run to backend with 90.0 second timeout
+2025-12-26 18:06:14,110 INFO    MainThread:189374 [wandb_init.py:init():1041] starting run threads in backend
+2025-12-26 18:06:14,225 INFO    MainThread:189374 [wandb_run.py:_console_start():2521] atexit reg
+2025-12-26 18:06:14,226 INFO    MainThread:189374 [wandb_run.py:_redirect():2369] redirect: wrap_raw
+2025-12-26 18:06:14,226 INFO    MainThread:189374 [wandb_run.py:_redirect():2438] Wrapping output streams.
+2025-12-26 18:06:14,226 INFO    MainThread:189374 [wandb_run.py:_redirect():2461] Redirects installed.
+2025-12-26 18:06:14,230 INFO    MainThread:189374 [wandb_init.py:init():1081] run started, returning control to user process
+2025-12-26 18:06:33,265 INFO    wandb-AsyncioManager-main:189374 [service_client.py:_forward_responses():80] Reached EOF.
+2025-12-26 18:06:33,265 INFO    wandb-AsyncioManager-main:189374 [mailbox.py:close():137] Closing mailbox, abandoning 1 handles.

wandb/run-20251226_180702-oordmylf/files/config.yaml

@@ -0,0 +1,173 @@
+_wandb:
+    value:
+        cli_version: 0.23.1
+        e:
+            b1hui6k7d05xwq8cyz4bv453ig8gyf1q:
+                args:
+                    - --config
+                    - trainer-kit/SFT/config_instruct.yaml
+                codePath: trainer-kit/SFT/run_instruct.py
+                codePathLocal: trainer-kit/SFT/run_instruct.py
+                cpu_count: 12
+                cpu_count_logical: 24
+                cudaVersion: "13.0"
+                disk:
+                    /:
+                        total: "791251738624"
+                        used: "386726633472"
+                email: shaiksirajuddin9949@gmail.com
+                executable: /workspace/llm_finetuning_env/bin/python
+                gpu: NVIDIA A100-SXM4-80GB
+                gpu_count: 2
+                gpu_nvidia:
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40
+                host: a100-2gpu-shell-session-757d587799-mfdvv
+                memory:
+                    total: "359047892992"
+                os: Linux-6.12.46+-x86_64-with-glibc2.35
+                program: /workspace/trainer-kit/SFT/run_instruct.py
+                python: CPython 3.10.12
+                root: task2file/sft_devstral_24B_v2
+                startedAt: "2025-12-26T18:07:02.184185Z"
+                writerId: b1hui6k7d05xwq8cyz4bv453ig8gyf1q
+        m: []
+        python_version: 3.10.12
+        t:
+            "1":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "2":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "3":
+                - 15
+                - 16
+            "4": 3.10.12
+            "5": 0.23.1
+            "6": 5.0.0.dev0
+            "12": 0.23.1
+            "13": linux-x86_64
+data:
+    value:
+        custom_template: |-
+            ##INSTRUCTION
+            {instruction}<|im_end|>
+            {input}<|im_end|>
+            {output}<|im_end|>
+        eval_jsonl: null
+        eval_split_ratio: 0.1
+        format_type: custom
+        input_field: input
+        instruction_field: instruction
+        max_length: 2048
+        num_proc: 4
+        output_field: output
+        shuffle: true
+        system_prompt: |
+            You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.
+
+            ## Output Format
+
+            ##OUTPUT
+            Explain the data flow and why each component must change:
+            - Flow: [Input → Processing → Output with arrows]
+            - For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"
+            - Explain coupling between components
+
+            ##SELECT
+            modify::crates/path/to/file.rs::impl::ComponentName
+            add::crates/another/file.rs::function::AnotherComponent
+            <EOS>
+
+            ## Rules
+
+            1. Use full paths: `remove::crates/folder/file.rs::Type::Name`
+            2. Use `::` for nested items: `status::StructName::Type::Name`
+            3. Always explain "must change because" and "without this"
+            3. Types of components: function, struct, enum, impl, trait
+            4. If there is extra information (e.g., enum variants), include that too.
+            5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>
+
+            ## Example
+
+            ##TASK
+            Add webhook subscription support
+
+            ##OUTPUT
+            The webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don't trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.
+
+            ##SELECT
+            crates/common_enums/src/enums.rs::EventClass
+            crates/common_enums/src/transformers.rs::SubscriptionStatus
+            <EOS>
+        train_jsonl: sft_dataset.jsonl
+model:
+    value:
+        attn_implementation: null
+        base_local_dir: base_model
+        bnb_4bit_compute_dtype: bfloat16
+        bnb_4bit_quant_type: nf4
+        bnb_4bit_use_double_quant: false
+        device_map: auto
+        repo_id: ./Models/Devstral-Small-2-24B-HS-CPT
+        revision: null
+        tokenizer_use_fast: true
+        torch_dtype: bfloat16
+        trust_remote_code: true
+        use_4bit: false
+peft:
+    value:
+        bias: none
+        enabled: true
+        lora_alpha: 16
+        lora_dropout: 0.05
+        r: 8
+        target_modules: auto
+run_dir:
+    value: task2file/sft_devstral_24B_v2
+train:
+    value:
+        early_stopping:
+            enabled: true
+            metric: eval_loss
+            min_delta: 0.001
+            mode: min
+            patience: 5
+        eval_steps: 100
+        evaluation_strategy: steps
+        gradient_accumulation_steps: 8
+        gradient_checkpointing: true
+        learning_rate: "1e-4"
+        load_best_model_at_end: true
+        logging_steps: 2
+        lr_scheduler_type: cosine
+        max_grad_norm: 0.8
+        num_train_epochs: 6
+        optim: adamw_torch
+        per_device_eval_batch_size: 1
+        per_device_train_batch_size: 1
+        resume_from_checkpoint: auto
+        save_steps: 500
+        save_strategy: steps
+        save_total_limit: 20
+        warmup_ratio: 0.08
+        weight_decay: 0

wandb/run-20251226_180702-oordmylf/files/output.log

@@ -0,0 +1,48 @@
+Wandb initialized: project='sft-training', name='auto-generated'
+[info] Detected Mistral3 model architecture, loading with specific class
+Loading weights: 100%|█| 585/585 [00:12<00:00, 47.03it/s, Materializing param=model.vision_tower.transfor
+[info] Ensuring all parameters are materialized...
+Traceback (most recent call last):
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 983, in <module>
+    main()
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 852, in main
+    train_ds, eval_ds = build_datasets(cfg, tokenizer)
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 414, in build_datasets
+    ds = load_dataset("json", data_files={"train": train_path})
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 1492, in load_dataset
+    builder_instance = load_dataset_builder(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 1137, in load_dataset_builder
+    dataset_module = dataset_module_factory(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 913, in dataset_module_factory
+    ).get_module()
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 527, in get_module
+    data_files = DataFilesDict.from_patterns(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 708, in from_patterns
+    else DataFilesList.from_patterns(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 601, in from_patterns
+    resolve_pattern(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 390, in resolve_pattern
+    raise FileNotFoundError(error_msg)
+FileNotFoundError: Unable to find '/workspace/sft_dataset.jsonl'
+Traceback (most recent call last):
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 983, in <module>
+    main()
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 852, in main
+    train_ds, eval_ds = build_datasets(cfg, tokenizer)
+  File "/workspace/trainer-kit/SFT/run_instruct.py", line 414, in build_datasets
+    ds = load_dataset("json", data_files={"train": train_path})
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 1492, in load_dataset
+    builder_instance = load_dataset_builder(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 1137, in load_dataset_builder
+    dataset_module = dataset_module_factory(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 913, in dataset_module_factory
+    ).get_module()
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/load.py", line 527, in get_module
+    data_files = DataFilesDict.from_patterns(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 708, in from_patterns
+    else DataFilesList.from_patterns(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 601, in from_patterns
+    resolve_pattern(
+  File "/workspace/llm_finetuning_env/lib/python3.10/site-packages/datasets/data_files.py", line 390, in resolve_pattern
+    raise FileNotFoundError(error_msg)
+FileNotFoundError: Unable to find '/workspace/sft_dataset.jsonl'

wandb/run-20251226_180702-oordmylf/files/requirements.txt

@@ -0,0 +1,104 @@
+exceptiongroup==1.3.1
+wheel==0.45.1
+python-dateutil==2.9.0.post0
+nvidia-ml-py==13.580.82
+huggingface_hub==1.2.3
+idna==3.11
+click==8.3.1
+numpy==2.2.6
+httpx==0.28.1
+tokenizers==0.22.1
+sympy==1.13.1
+yarl==1.22.0
+async-timeout==5.0.1
+datasets==4.4.2
+platformdirs==4.5.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-nvtx-cu12==12.1.105
+smmap==5.0.2
+accelerate==1.12.0
+requests==2.32.5
+aiohttp==3.13.2
+bitsandbytes==0.49.0
+nvidia-cublas-cu12==12.1.3.1
+mpmath==1.3.0
+torchaudio==2.5.1+cu121
+nvidia-cuda-runtime-cu12==12.1.105
+typing-inspection==0.4.2
+GitPython==3.1.45
+xxhash==3.6.0
+nvidia-cusolver-cu12==11.4.5.107
+pydantic_core==2.41.5
+six==1.17.0
+torchvision==0.20.1+cu121
+typing_extensions==4.15.0
+triton==3.1.0
+charset-normalizer==3.4.4
+nvitop==1.6.1
+wandb==0.23.1
+regex==2025.11.3
+pip==25.3
+nvidia-cusparse-cu12==12.1.0.106
+pytz==2025.2
+Jinja2==3.1.6
+psutil==7.2.0
+pillow==12.0.0
+packaging==25.0
+safetensors==0.7.0
+sentry-sdk==2.48.0
+gitdb==4.0.12
+httpcore==1.0.9
+setuptools==80.9.0
+nvidia-cufft-cu12==11.0.2.54
+anyio==4.12.0
+transformers==5.0.0.dev0
+pydantic==2.12.5
+fsspec==2025.10.0
+filelock==3.20.0
+PyYAML==6.0.3
+hf-xet==1.2.0
+nvidia-cudnn-cu12==9.1.0.70
+tqdm==4.67.1
+MarkupSafe==2.1.5
+attrs==25.4.0
+nvidia-cuda-nvrtc-cu12==12.1.105
+peft==0.18.0
+aiohappyeyeballs==2.6.1
+networkx==3.4.2
+nvidia-nvjitlink-cu12==12.9.86
+certifi==2025.11.12
+pyarrow==22.0.0
+dill==0.4.0
+protobuf==6.33.2
+aiosignal==1.4.0
+frozenlist==1.8.0
+urllib3==2.6.2
+propcache==0.4.1
+tzdata==2025.3
+pandas==2.3.3
+annotated-types==0.7.0
+shellingham==1.5.4
+nvidia-nccl-cu12==2.21.5
+multidict==6.7.0
+nvidia-curand-cu12==10.3.2.106
+trl==0.26.2
+torch==2.5.1+cu121
+h11==0.16.0
+multiprocess==0.70.18
+typer-slim==0.21.0
+wheel==0.45.1
+tomli==2.0.1
+autocommand==2.2.2
+jaraco.context==5.3.0
+zipp==3.19.2
+packaging==24.2
+inflect==7.3.1
+typing_extensions==4.12.2
+platformdirs==4.2.2
+jaraco.functools==4.0.1
+jaraco.collections==5.1.0
+jaraco.text==3.12.1
+backports.tarfile==1.2.0
+more-itertools==10.3.0
+importlib_metadata==8.0.0
+typeguard==4.3.0

wandb/run-20251226_180702-oordmylf/files/wandb-metadata.json

@@ -0,0 +1,47 @@
+{
+  "os":  "Linux-6.12.46+-x86_64-with-glibc2.35",
+  "python":  "CPython 3.10.12",
+  "startedAt":  "2025-12-26T18:07:02.184185Z",
+  "args":  [
+    "--config",
+    "trainer-kit/SFT/config_instruct.yaml"
+  ],
+  "program":  "/workspace/trainer-kit/SFT/run_instruct.py",
+  "codePath":  "trainer-kit/SFT/run_instruct.py",
+  "codePathLocal":  "trainer-kit/SFT/run_instruct.py",
+  "email":  "shaiksirajuddin9949@gmail.com",
+  "root":  "task2file/sft_devstral_24B_v2",
+  "host":  "a100-2gpu-shell-session-757d587799-mfdvv",
+  "executable":  "/workspace/llm_finetuning_env/bin/python",
+  "cpu_count":  12,
+  "cpu_count_logical":  24,
+  "gpu":  "NVIDIA A100-SXM4-80GB",
+  "gpu_count":  2,
+  "disk":  {
+    "/":  {
+      "total":  "791251738624",
+      "used":  "386726633472"
+    }
+  },
+  "memory":  {
+    "total":  "359047892992"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba"
+    },
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40"
+    }
+  ],
+  "cudaVersion":  "13.0",
+  "writerId":  "b1hui6k7d05xwq8cyz4bv453ig8gyf1q"
+}

wandb/run-20251226_180702-oordmylf/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_wandb":{"runtime":19},"_runtime":19}

wandb/run-20251226_180702-oordmylf/logs/debug-core.log

@@ -0,0 +1,14 @@
+{"time":"2025-12-26T18:07:02.269040435Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpn17hlpsl/port-189808.txt","pid":189808,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2025-12-26T18:07:02.269706147Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":189808}
+{"time":"2025-12-26T18:07:02.26970836Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-189808-189898-3164802245/socket","Net":"unix"}}
+{"time":"2025-12-26T18:07:02.451619132Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2025-12-26T18:07:02.458183004Z","level":"INFO","msg":"handleInformInit: received","streamId":"oordmylf","id":"1(@)"}
+{"time":"2025-12-26T18:07:02.645894292Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"oordmylf","id":"1(@)"}
+{"time":"2025-12-26T18:07:21.83310619Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2025-12-26T18:07:21.833169189Z","level":"INFO","msg":"server is shutting down"}
+{"time":"2025-12-26T18:07:21.833173516Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2025-12-26T18:07:21.833268292Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-189808-189898-3164802245/socket","Net":"unix"}}
+{"time":"2025-12-26T18:07:21.833292467Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2025-12-26T18:07:22.158467464Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2025-12-26T18:07:22.158501256Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2025-12-26T18:07:22.158518889Z","level":"INFO","msg":"server is closed"}

wandb/run-20251226_180702-oordmylf/logs/debug-internal.log

@@ -0,0 +1,11 @@
+{"time":"2025-12-26T18:07:02.458313987Z","level":"INFO","msg":"stream: starting","core version":"0.23.1"}
+{"time":"2025-12-26T18:07:02.645556611Z","level":"INFO","msg":"stream: created new stream","id":"oordmylf"}
+{"time":"2025-12-26T18:07:02.645689372Z","level":"INFO","msg":"handler: started","stream_id":"oordmylf"}
+{"time":"2025-12-26T18:07:02.645880522Z","level":"INFO","msg":"stream: started","id":"oordmylf"}
+{"time":"2025-12-26T18:07:02.64591008Z","level":"INFO","msg":"writer: started","stream_id":"oordmylf"}
+{"time":"2025-12-26T18:07:02.64593173Z","level":"INFO","msg":"sender: started","stream_id":"oordmylf"}
+{"time":"2025-12-26T18:07:21.833167126Z","level":"INFO","msg":"stream: closing","id":"oordmylf"}
+{"time":"2025-12-26T18:07:22.023419324Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2025-12-26T18:07:22.157584616Z","level":"INFO","msg":"handler: closed","stream_id":"oordmylf"}
+{"time":"2025-12-26T18:07:22.157683147Z","level":"INFO","msg":"sender: closed","stream_id":"oordmylf"}
+{"time":"2025-12-26T18:07:22.157690597Z","level":"INFO","msg":"stream: closed","id":"oordmylf"}

wandb/run-20251226_180702-oordmylf/logs/debug.log

@@ -0,0 +1,23 @@
+2025-12-26 18:07:02,185 INFO    MainThread:189808 [wandb_setup.py:_flush():80] Current SDK version is 0.23.1
+2025-12-26 18:07:02,185 INFO    MainThread:189808 [wandb_setup.py:_flush():80] Configure stats pid to 189808
+2025-12-26 18:07:02,186 INFO    MainThread:189808 [wandb_setup.py:_flush():80] Loading settings from /root/.config/wandb/settings
+2025-12-26 18:07:02,186 INFO    MainThread:189808 [wandb_setup.py:_flush():80] Loading settings from /workspace/wandb/settings
+2025-12-26 18:07:02,186 INFO    MainThread:189808 [wandb_setup.py:_flush():80] Loading settings from environment variables
+2025-12-26 18:07:02,186 INFO    MainThread:189808 [wandb_init.py:setup_run_log_directory():714] Logging user logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180702-oordmylf/logs/debug.log
+2025-12-26 18:07:02,186 INFO    MainThread:189808 [wandb_init.py:setup_run_log_directory():715] Logging internal logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180702-oordmylf/logs/debug-internal.log
+2025-12-26 18:07:02,186 INFO    MainThread:189808 [wandb_init.py:init():841] calling init triggers
+2025-12-26 18:07:02,186 INFO    MainThread:189808 [wandb_init.py:init():846] wandb.init called with sweep_config: {}
+config: {'model': {'repo_id': './Models/Devstral-Small-2-24B-HS-CPT', 'revision': None, 'base_local_dir': 'base_model', 'trust_remote_code': True, 'tokenizer_use_fast': True, 'device_map': 'auto', 'torch_dtype': 'bfloat16', 'use_4bit': False, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': False, 'bnb_4bit_compute_dtype': 'bfloat16', 'attn_implementation': None}, 'data': {'train_jsonl': 'sft_dataset.jsonl', 'eval_jsonl': None, 'eval_split_ratio': 0.1, 'instruction_field': 'instruction', 'input_field': 'input', 'output_field': 'output', 'format_type': 'custom', 'system_prompt': 'You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain the data flow and why each component must change:\n- Flow: [Input → Processing → Output with arrows]\n- For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"\n- Explain coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\nadd::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for nested items: `status::StructName::Type::Name`\n3. Always explain "must change because" and "without this"\n3. Types of components: function, struct, enum, impl, trait\n4. If there is extra information (e.g., enum variants), include that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don\'t trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\ncrates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n', 'custom_template': '##INSTRUCTION\n{instruction}<|im_end|>\n{input}<|im_end|>\n{output}<|im_end|>', 'max_length': 2048, 'shuffle': True, 'num_proc': 4}, 'peft': {'enabled': True, 'r': 8, 'lora_alpha': 16, 'lora_dropout': 0.05, 'bias': 'none', 'target_modules': 'auto'}, 'train': {'num_train_epochs': 6, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'learning_rate': '1e-4', 'weight_decay': 0.0, 'warmup_ratio': 0.08, 'lr_scheduler_type': 'cosine', 'optim': 'adamw_torch', 'max_grad_norm': 0.8, 'gradient_checkpointing': True, 'logging_steps': 2, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'evaluation_strategy': 'steps', 'eval_steps': 100, 'load_best_model_at_end': True, 'early_stopping': {'enabled': True, 'patience': 5, 'min_delta': 0.001, 'metric': 'eval_loss', 'mode': 'min'}, 'resume_from_checkpoint': 'auto'}, 'run_dir': 'task2file/sft_devstral_24B_v2', '_wandb': {}}
+2025-12-26 18:07:02,186 INFO    MainThread:189808 [wandb_init.py:init():889] starting backend
+2025-12-26 18:07:02,451 INFO    MainThread:189808 [wandb_init.py:init():892] sending inform_init request
+2025-12-26 18:07:02,456 INFO    MainThread:189808 [wandb_init.py:init():900] backend started and connected
+2025-12-26 18:07:02,459 INFO    MainThread:189808 [wandb_init.py:init():970] updated telemetry
+2025-12-26 18:07:02,460 INFO    MainThread:189808 [wandb_init.py:init():994] communicating run to backend with 90.0 second timeout
+2025-12-26 18:07:02,828 INFO    MainThread:189808 [wandb_init.py:init():1041] starting run threads in backend
+2025-12-26 18:07:02,938 INFO    MainThread:189808 [wandb_run.py:_console_start():2521] atexit reg
+2025-12-26 18:07:02,938 INFO    MainThread:189808 [wandb_run.py:_redirect():2369] redirect: wrap_raw
+2025-12-26 18:07:02,938 INFO    MainThread:189808 [wandb_run.py:_redirect():2438] Wrapping output streams.
+2025-12-26 18:07:02,938 INFO    MainThread:189808 [wandb_run.py:_redirect():2461] Redirects installed.
+2025-12-26 18:07:02,942 INFO    MainThread:189808 [wandb_init.py:init():1081] run started, returning control to user process
+2025-12-26 18:07:21,833 INFO    wandb-AsyncioManager-main:189808 [service_client.py:_forward_responses():80] Reached EOF.
+2025-12-26 18:07:21,833 INFO    wandb-AsyncioManager-main:189808 [mailbox.py:close():137] Closing mailbox, abandoning 1 handles.

wandb/run-20251226_180808-ny9q48hd/files/config.yaml

@@ -0,0 +1,630 @@
+_name_or_path:
+    value: Models/Devstral-Small-2-24B-HS-CPT
+_wandb:
+    value:
+        cli_version: 0.23.1
+        e:
+            k9a7zk6glrk7p1134kg1guh8b7acqlob:
+                args:
+                    - --config
+                    - trainer-kit/SFT/config_instruct.yaml
+                codePath: trainer-kit/SFT/run_instruct.py
+                codePathLocal: trainer-kit/SFT/run_instruct.py
+                cpu_count: 12
+                cpu_count_logical: 24
+                cudaVersion: "13.0"
+                disk:
+                    /:
+                        total: "791251738624"
+                        used: "387681259520"
+                email: shaiksirajuddin9949@gmail.com
+                executable: /workspace/llm_finetuning_env/bin/python
+                gpu: NVIDIA A100-SXM4-80GB
+                gpu_count: 2
+                gpu_nvidia:
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba
+                    - architecture: Ampere
+                      cudaCores: 6912
+                      memoryTotal: "85899345920"
+                      name: NVIDIA A100-SXM4-80GB
+                      uuid: GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40
+                host: a100-2gpu-shell-session-757d587799-mfdvv
+                memory:
+                    total: "359047892992"
+                os: Linux-6.12.46+-x86_64-with-glibc2.35
+                program: /workspace/trainer-kit/SFT/run_instruct.py
+                python: CPython 3.10.12
+                root: task2file/sft_devstral_24B_v2
+                startedAt: "2025-12-26T18:08:08.383305Z"
+                writerId: k9a7zk6glrk7p1134kg1guh8b7acqlob
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.10.12
+        t:
+            "1":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "2":
+                - 1
+                - 11
+                - 41
+                - 49
+                - 51
+                - 71
+                - 98
+            "3":
+                - 2
+                - 7
+                - 15
+                - 16
+                - 19
+                - 62
+                - 66
+            "4": 3.10.12
+            "5": 0.23.1
+            "6": 5.0.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.23.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+add_cross_attention:
+    value: false
+architectures:
+    value:
+        - Mistral3ForConditionalGeneration
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: null
+chunk_size_feed_forward:
+    value: 0
+cross_attention_hidden_size:
+    value: null
+data:
+    value:
+        custom_template: |-
+            ##INSTRUCTION
+            {instruction}<|im_end|>
+            {input}<|im_end|>
+            {output}<|im_end|>
+        eval_jsonl: null
+        eval_split_ratio: 0.1
+        format_type: custom
+        input_field: input
+        instruction_field: instruction
+        max_length: 2048
+        num_proc: 4
+        output_field: output
+        shuffle: true
+        system_prompt: |
+            You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.
+
+            ## Output Format
+
+            ##OUTPUT
+            Explain the data flow and why each component must change:
+            - Flow: [Input → Processing → Output with arrows]
+            - For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"
+            - Explain coupling between components
+
+            ##SELECT
+            modify::crates/path/to/file.rs::impl::ComponentName
+            add::crates/another/file.rs::function::AnotherComponent
+            <EOS>
+
+            ## Rules
+
+            1. Use full paths: `remove::crates/folder/file.rs::Type::Name`
+            2. Use `::` for nested items: `status::StructName::Type::Name`
+            3. Always explain "must change because" and "without this"
+            3. Types of components: function, struct, enum, impl, trait
+            4. If there is extra information (e.g., enum variants), include that too.
+            5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>
+
+            ## Example
+
+            ##TASK
+            Add webhook subscription support
+
+            ##OUTPUT
+            The webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don't trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.
+
+            ##SELECT
+            crates/common_enums/src/enums.rs::EventClass
+            crates/common_enums/src/transformers.rs::SubscriptionStatus
+            <EOS>
+        train_jsonl: sft_dataset.jsonl
+data_seed:
+    value: null
+dataloader_drop_last:
+    value: false
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+decoder_start_token_id:
+    value: null
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: bfloat16
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: null
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 100
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+finetuning_task:
+    value: null
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: []
+fsdp_config:
+    value:
+        min_num_params: 0
+        xla: false
+        xla_fsdp_grad_ckpt: false
+        xla_fsdp_v2: false
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 8
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: false
+group_by_length:
+    value: false
+hub_always_push:
+    value: false
+hub_model_id:
+    value: null
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+image_token_index:
+    value: 10
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+is_decoder:
+    value: false
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.0001
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: true
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_dir:
+    value: null
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 2
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: cosine
+max_grad_norm:
+    value: 0.8
+max_steps:
+    value: -1
+metric_for_best_model:
+    value: eval_loss
+model:
+    value:
+        attn_implementation: null
+        base_local_dir: base_model
+        bnb_4bit_compute_dtype: bfloat16
+        bnb_4bit_quant_type: nf4
+        bnb_4bit_use_double_quant: false
+        device_map: auto
+        repo_id: ./Models/Devstral-Small-2-24B-HS-CPT
+        revision: null
+        tokenizer_use_fast: true
+        torch_dtype: bfloat16
+        trust_remote_code: true
+        use_4bit: false
+model/num_parameters:
+    value: 24022764544
+model_type:
+    value: mistral3
+multimodal_projector_bias:
+    value: false
+neftune_noise_alpha:
+    value: null
+num_train_epochs:
+    value: 6
+optim:
+    value: adamw_torch
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: task2file/sft_devstral_24B_v2/checkpoints
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: null
+parallelism_config:
+    value: null
+peft:
+    value:
+        bias: none
+        enabled: true
+        lora_alpha: 16
+        lora_dropout: 0.05
+        r: 8
+        target_modules: auto
+peft_config:
+    value:
+        default:
+            alora_invocation_tokens: null
+            arrow_config: null
+            auto_mapping: null
+            base_model_name_or_path: Models/Devstral-Small-2-24B-HS-CPT
+            bias: none
+            corda_config: null
+            ensure_weight_tying: false
+            eva_config: null
+            exclude_modules: null
+            fan_in_fan_out: false
+            inference_mode: false
+            init_lora_weights: true
+            layer_replication: null
+            layers_pattern: null
+            layers_to_transform: null
+            lora_alpha: 16
+            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: 8
+            revision: null
+            runtime_config:
+                ephemeral_gpu_offload: false
+            target_modules:
+                - v_proj
+                - q_proj
+                - o_proj
+                - k_proj
+            target_parameters: null
+            task_type: CAUSAL_LM
+            trainable_token_indices: null
+            use_dora: false
+            use_qalora: false
+            use_rslora: false
+per_device_eval_batch_size:
+    value: 1
+per_device_train_batch_size:
+    value: 1
+prediction_loss_only:
+    value: false
+prefix:
+    value: null
+problem_type:
+    value: null
+project:
+    value: huggingface
+projector_hidden_act:
+    value: gelu
+push_to_hub:
+    value: false
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+run_dir:
+    value: task2file/sft_devstral_24B_v2
+run_name:
+    value: null
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 500
+save_strategy:
+    value: steps
+save_total_limit:
+    value: 20
+seed:
+    value: 42
+sep_token_id:
+    value: null
+skip_memory_metrics:
+    value: true
+spatial_merge_size:
+    value: 2
+task_specific_params:
+    value: null
+text_config:
+    value:
+        _name_or_path: ""
+        add_cross_attention: false
+        architectures: null
+        attention_dropout: 0
+        bos_token_id: 1
+        chunk_size_feed_forward: 0
+        cross_attention_hidden_size: null
+        decoder_start_token_id: null
+        dtype: bfloat16
+        eos_token_id: 2
+        finetuning_task: null
+        head_dim: 128
+        hidden_act: silu
+        hidden_size: 5120
+        id2label:
+            "0": LABEL_0
+            "1": LABEL_1
+        initializer_range: 0.02
+        intermediate_size: 32768
+        is_decoder: false
+        is_encoder_decoder: false
+        label2id:
+            LABEL_0: 0
+            LABEL_1: 1
+        max_position_embeddings: 393216
+        model_type: ministral3
+        num_attention_heads: 32
+        num_hidden_layers: 40
+        num_key_value_heads: 8
+        output_attentions: false
+        output_hidden_states: false
+        pad_token_id: 11
+        prefix: null
+        problem_type: null
+        return_dict: true
+        rms_norm_eps: 1e-05
+        rope_parameters:
+            beta_fast: 32
+            beta_slow: 1
+            factor: 48
+            llama_4_scaling_beta: 0.1
+            mscale: 1
+            mscale_all_dim: 1
+            original_max_position_embeddings: 8192
+            rope_theta: 1e+08
+            rope_type: yarn
+            type: yarn
+        sep_token_id: null
+        sliding_window: null
+        task_specific_params: null
+        tie_word_embeddings: false
+        tokenizer_class: null
+        use_cache: true
+        vocab_size: 131072
+tf32:
+    value: null
+tie_word_embeddings:
+    value: false
+tokenizer_class:
+    value: null
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_space_id:
+    value: trackio
+train:
+    value:
+        early_stopping:
+            enabled: true
+            metric: eval_loss
+            min_delta: 0.001
+            mode: min
+            patience: 5
+        eval_steps: 100
+        evaluation_strategy: steps
+        gradient_accumulation_steps: 8
+        gradient_checkpointing: true
+        learning_rate: "1e-4"
+        load_best_model_at_end: true
+        logging_steps: 2
+        lr_scheduler_type: cosine
+        max_grad_norm: 0.8
+        num_train_epochs: 6
+        optim: adamw_torch
+        per_device_eval_batch_size: 1
+        per_device_train_batch_size: 1
+        resume_from_checkpoint: auto
+        save_steps: 500
+        save_strategy: steps
+        save_total_limit: 20
+        warmup_ratio: 0.08
+        weight_decay: 0
+transformers_version:
+    value: 5.0.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vision_config:
+    value:
+        _name_or_path: ""
+        add_cross_attention: false
+        architectures: null
+        attention_dropout: 0
+        bos_token_id: null
+        chunk_size_feed_forward: 0
+        cross_attention_hidden_size: null
+        decoder_start_token_id: null
+        dtype: bfloat16
+        eos_token_id: null
+        finetuning_task: null
+        head_dim: 64
+        hidden_act: silu
+        hidden_size: 1024
+        id2label:
+            "0": LABEL_0
+            "1": LABEL_1
+        image_size: 1540
+        initializer_range: 0.02
+        intermediate_size: 4096
+        is_decoder: false
+        is_encoder_decoder: false
+        label2id:
+            LABEL_0: 0
+            LABEL_1: 1
+        model_type: pixtral
+        num_attention_heads: 16
+        num_channels: 3
+        num_hidden_layers: 24
+        output_attentions: false
+        output_hidden_states: false
+        pad_token_id: null
+        patch_size: 14
+        prefix: null
+        problem_type: null
+        return_dict: true
+        rope_parameters:
+            rope_theta: 10000
+            rope_type: default
+        sep_token_id: null
+        task_specific_params: null
+        tie_word_embeddings: true
+        tokenizer_class: null
+vision_feature_layer:
+    value: -1
+warmup_ratio:
+    value: 0.08
+warmup_steps:
+    value: 0.08
+weight_decay:
+    value: 0

wandb/run-20251226_180808-ny9q48hd/files/requirements.txt

@@ -0,0 +1,104 @@
+exceptiongroup==1.3.1
+wheel==0.45.1
+python-dateutil==2.9.0.post0
+nvidia-ml-py==13.580.82
+huggingface_hub==1.2.3
+idna==3.11
+click==8.3.1
+numpy==2.2.6
+httpx==0.28.1
+tokenizers==0.22.1
+sympy==1.13.1
+yarl==1.22.0
+async-timeout==5.0.1
+datasets==4.4.2
+platformdirs==4.5.1
+nvidia-cuda-cupti-cu12==12.1.105
+nvidia-nvtx-cu12==12.1.105
+smmap==5.0.2
+accelerate==1.12.0
+requests==2.32.5
+aiohttp==3.13.2
+bitsandbytes==0.49.0
+nvidia-cublas-cu12==12.1.3.1
+mpmath==1.3.0
+torchaudio==2.5.1+cu121
+nvidia-cuda-runtime-cu12==12.1.105
+typing-inspection==0.4.2
+GitPython==3.1.45
+xxhash==3.6.0
+nvidia-cusolver-cu12==11.4.5.107
+pydantic_core==2.41.5
+six==1.17.0
+torchvision==0.20.1+cu121
+typing_extensions==4.15.0
+triton==3.1.0
+charset-normalizer==3.4.4
+nvitop==1.6.1
+wandb==0.23.1
+regex==2025.11.3
+pip==25.3
+nvidia-cusparse-cu12==12.1.0.106
+pytz==2025.2
+Jinja2==3.1.6
+psutil==7.2.0
+pillow==12.0.0
+packaging==25.0
+safetensors==0.7.0
+sentry-sdk==2.48.0
+gitdb==4.0.12
+httpcore==1.0.9
+setuptools==80.9.0
+nvidia-cufft-cu12==11.0.2.54
+anyio==4.12.0
+transformers==5.0.0.dev0
+pydantic==2.12.5
+fsspec==2025.10.0
+filelock==3.20.0
+PyYAML==6.0.3
+hf-xet==1.2.0
+nvidia-cudnn-cu12==9.1.0.70
+tqdm==4.67.1
+MarkupSafe==2.1.5
+attrs==25.4.0
+nvidia-cuda-nvrtc-cu12==12.1.105
+peft==0.18.0
+aiohappyeyeballs==2.6.1
+networkx==3.4.2
+nvidia-nvjitlink-cu12==12.9.86
+certifi==2025.11.12
+pyarrow==22.0.0
+dill==0.4.0
+protobuf==6.33.2
+aiosignal==1.4.0
+frozenlist==1.8.0
+urllib3==2.6.2
+propcache==0.4.1
+tzdata==2025.3
+pandas==2.3.3
+annotated-types==0.7.0
+shellingham==1.5.4
+nvidia-nccl-cu12==2.21.5
+multidict==6.7.0
+nvidia-curand-cu12==10.3.2.106
+trl==0.26.2
+torch==2.5.1+cu121
+h11==0.16.0
+multiprocess==0.70.18
+typer-slim==0.21.0
+wheel==0.45.1
+tomli==2.0.1
+autocommand==2.2.2
+jaraco.context==5.3.0
+zipp==3.19.2
+packaging==24.2
+inflect==7.3.1
+typing_extensions==4.12.2
+platformdirs==4.2.2
+jaraco.functools==4.0.1
+jaraco.collections==5.1.0
+jaraco.text==3.12.1
+backports.tarfile==1.2.0
+more-itertools==10.3.0
+importlib_metadata==8.0.0
+typeguard==4.3.0

wandb/run-20251226_180808-ny9q48hd/files/wandb-metadata.json

@@ -0,0 +1,47 @@
+{
+  "os":  "Linux-6.12.46+-x86_64-with-glibc2.35",
+  "python":  "CPython 3.10.12",
+  "startedAt":  "2025-12-26T18:08:08.383305Z",
+  "args":  [
+    "--config",
+    "trainer-kit/SFT/config_instruct.yaml"
+  ],
+  "program":  "/workspace/trainer-kit/SFT/run_instruct.py",
+  "codePath":  "trainer-kit/SFT/run_instruct.py",
+  "codePathLocal":  "trainer-kit/SFT/run_instruct.py",
+  "email":  "shaiksirajuddin9949@gmail.com",
+  "root":  "task2file/sft_devstral_24B_v2",
+  "host":  "a100-2gpu-shell-session-757d587799-mfdvv",
+  "executable":  "/workspace/llm_finetuning_env/bin/python",
+  "cpu_count":  12,
+  "cpu_count_logical":  24,
+  "gpu":  "NVIDIA A100-SXM4-80GB",
+  "gpu_count":  2,
+  "disk":  {
+    "/":  {
+      "total":  "791251738624",
+      "used":  "387681259520"
+    }
+  },
+  "memory":  {
+    "total":  "359047892992"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-989794b0-ec3b-13bf-db9f-3fbe341497ba"
+    },
+    {
+      "name":  "NVIDIA A100-SXM4-80GB",
+      "memoryTotal":  "85899345920",
+      "cudaCores":  6912,
+      "architecture":  "Ampere",
+      "uuid":  "GPU-3790aa64-60ef-9eac-b0b1-b278ee8c0d40"
+    }
+  ],
+  "cudaVersion":  "13.0",
+  "writerId":  "k9a7zk6glrk7p1134kg1guh8b7acqlob"
+}

wandb/run-20251226_180808-ny9q48hd/files/wandb-summary.json

@@ -0,0 +1 @@
+{"eval/runtime":511.6513,"_timestamp":1.766894424089547e+09,"train/learning_rate":5.0960525730763455e-05,"train_runtime":121379.3179,"_wandb":{"runtime":122035},"eval/samples_per_second":4.118,"train_loss":0.6813117427967097,"total_flos":7.892056292508187e+18,"train/loss":0.5559571981430054,"eval/steps_per_second":4.118,"_step":3877,"train/global_step":7600,"train_samples_per_second":0.937,"train/grad_norm":1.2778210639953613,"train_steps_per_second":0.117,"eval/loss":0.6706293225288391,"_runtime":122035,"train/epoch":3.2067510548523206}

wandb/run-20251226_180808-ny9q48hd/logs/debug-core.log

@@ -0,0 +1,16 @@
+{"time":"2025-12-26T18:08:08.471552693Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp_2pxhrkk/port-190322.txt","pid":190322,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2025-12-26T18:08:08.472277502Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":190322}
+{"time":"2025-12-26T18:08:08.472253441Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-190322-190397-1549185738/socket","Net":"unix"}}
+{"time":"2025-12-26T18:08:08.653761101Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2025-12-26T18:08:08.660887612Z","level":"INFO","msg":"handleInformInit: received","streamId":"ny9q48hd","id":"1(@)"}
+{"time":"2025-12-26T18:08:08.822064564Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"ny9q48hd","id":"1(@)"}
+{"time":"2025-12-28T04:02:05.051125329Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"ny9q48hd","id":"1(@)"}
+{"time":"2025-12-28T04:02:05.052538266Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"ny9q48hd","id":"1(@)"}
+{"time":"2025-12-28T04:02:05.107259931Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2025-12-28T04:02:05.107301964Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2025-12-28T04:02:05.107312563Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2025-12-28T04:02:05.107351058Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2025-12-28T04:02:05.107355378Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2025-12-28T04:02:05.107365658Z","level":"INFO","msg":"server is shutting down"}
+{"time":"2025-12-28T04:02:05.107515239Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-190322-190397-1549185738/socket","Net":"unix"}}
+{"time":"2025-12-28T04:02:05.107566032Z","level":"INFO","msg":"server is closed"}

wandb/run-20251226_180808-ny9q48hd/logs/debug-internal.log

@@ -0,0 +1,12 @@
+{"time":"2025-12-26T18:08:08.66103332Z","level":"INFO","msg":"stream: starting","core version":"0.23.1"}
+{"time":"2025-12-26T18:08:08.82172381Z","level":"INFO","msg":"stream: created new stream","id":"ny9q48hd"}
+{"time":"2025-12-26T18:08:08.821819478Z","level":"INFO","msg":"handler: started","stream_id":"ny9q48hd"}
+{"time":"2025-12-26T18:08:08.822049155Z","level":"INFO","msg":"stream: started","id":"ny9q48hd"}
+{"time":"2025-12-26T18:08:08.822072296Z","level":"INFO","msg":"writer: started","stream_id":"ny9q48hd"}
+{"time":"2025-12-26T18:08:08.822098276Z","level":"INFO","msg":"sender: started","stream_id":"ny9q48hd"}
+{"time":"2025-12-28T04:02:04.935383596Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2025-12-28T04:02:05.045953421Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2025-12-28T04:02:05.051806259Z","level":"INFO","msg":"stream: closing","id":"ny9q48hd"}
+{"time":"2025-12-28T04:02:05.051833004Z","level":"INFO","msg":"handler: closed","stream_id":"ny9q48hd"}
+{"time":"2025-12-28T04:02:05.051917075Z","level":"INFO","msg":"sender: closed","stream_id":"ny9q48hd"}
+{"time":"2025-12-28T04:02:05.051937152Z","level":"INFO","msg":"stream: closed","id":"ny9q48hd"}

wandb/run-20251226_180808-ny9q48hd/logs/debug.log

@@ -0,0 +1,29 @@
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Current SDK version is 0.23.1
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Configure stats pid to 190322
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Loading settings from /root/.config/wandb/settings
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Loading settings from /workspace/wandb/settings
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_setup.py:_flush():80] Loading settings from environment variables
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:setup_run_log_directory():714] Logging user logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180808-ny9q48hd/logs/debug.log
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:setup_run_log_directory():715] Logging internal logs to task2file/sft_devstral_24B_v2/wandb/run-20251226_180808-ny9q48hd/logs/debug-internal.log
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:init():841] calling init triggers
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:init():846] wandb.init called with sweep_config: {}
+config: {'model': {'repo_id': './Models/Devstral-Small-2-24B-HS-CPT', 'revision': None, 'base_local_dir': 'base_model', 'trust_remote_code': True, 'tokenizer_use_fast': True, 'device_map': 'auto', 'torch_dtype': 'bfloat16', 'use_4bit': False, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': False, 'bnb_4bit_compute_dtype': 'bfloat16', 'attn_implementation': None}, 'data': {'train_jsonl': 'sft_dataset.jsonl', 'eval_jsonl': None, 'eval_split_ratio': 0.1, 'instruction_field': 'instruction', 'input_field': 'input', 'output_field': 'output', 'format_type': 'custom', 'system_prompt': 'You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain the data flow and why each component must change:\n- Flow: [Input → Processing → Output with arrows]\n- For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"\n- Explain coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\nadd::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for nested items: `status::StructName::Type::Name`\n3. Always explain "must change because" and "without this"\n3. Types of components: function, struct, enum, impl, trait\n4. If there is extra information (e.g., enum variants), include that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don\'t trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\ncrates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n', 'custom_template': '##INSTRUCTION\n{instruction}<|im_end|>\n{input}<|im_end|>\n{output}<|im_end|>', 'max_length': 2048, 'shuffle': True, 'num_proc': 4}, 'peft': {'enabled': True, 'r': 8, 'lora_alpha': 16, 'lora_dropout': 0.05, 'bias': 'none', 'target_modules': 'auto'}, 'train': {'num_train_epochs': 6, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'learning_rate': '1e-4', 'weight_decay': 0.0, 'warmup_ratio': 0.08, 'lr_scheduler_type': 'cosine', 'optim': 'adamw_torch', 'max_grad_norm': 0.8, 'gradient_checkpointing': True, 'logging_steps': 2, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'evaluation_strategy': 'steps', 'eval_steps': 100, 'load_best_model_at_end': True, 'early_stopping': {'enabled': True, 'patience': 5, 'min_delta': 0.001, 'metric': 'eval_loss', 'mode': 'min'}, 'resume_from_checkpoint': 'auto'}, 'run_dir': 'task2file/sft_devstral_24B_v2', '_wandb': {}}
+2025-12-26 18:08:08,385 INFO    MainThread:190322 [wandb_init.py:init():889] starting backend
+2025-12-26 18:08:08,653 INFO    MainThread:190322 [wandb_init.py:init():892] sending inform_init request
+2025-12-26 18:08:08,658 INFO    MainThread:190322 [wandb_init.py:init():900] backend started and connected
+2025-12-26 18:08:08,661 INFO    MainThread:190322 [wandb_init.py:init():970] updated telemetry
+2025-12-26 18:08:08,662 INFO    MainThread:190322 [wandb_init.py:init():994] communicating run to backend with 90.0 second timeout
+2025-12-26 18:08:09,021 INFO    MainThread:190322 [wandb_init.py:init():1041] starting run threads in backend
+2025-12-26 18:08:09,134 INFO    MainThread:190322 [wandb_run.py:_console_start():2521] atexit reg
+2025-12-26 18:08:09,134 INFO    MainThread:190322 [wandb_run.py:_redirect():2369] redirect: wrap_raw
+2025-12-26 18:08:09,135 INFO    MainThread:190322 [wandb_run.py:_redirect():2438] Wrapping output streams.
+2025-12-26 18:08:09,135 INFO    MainThread:190322 [wandb_run.py:_redirect():2461] Redirects installed.
+2025-12-26 18:08:09,138 INFO    MainThread:190322 [wandb_init.py:init():1081] run started, returning control to user process
+2025-12-26 18:08:52,955 INFO    MainThread:190322 [wandb_run.py:_config_callback():1396] config_cb None None {'peft_config': {'default': {'task_type': 'CAUSAL_LM', 'peft_type': 'LORA', 'auto_mapping': None, 'peft_version': '0.18.0', 'base_model_name_or_path': 'Models/Devstral-Small-2-24B-HS-CPT', 'revision': None, 'inference_mode': False, 'r': 8, 'target_modules': ['v_proj', 'q_proj', 'o_proj', 'k_proj'], 'exclude_modules': None, 'lora_alpha': 16, 'lora_dropout': 0.05, 'fan_in_fan_out': False, 'bias': 'none', 'use_rslora': False, 'modules_to_save': None, 'init_lora_weights': True, 'layers_to_transform': None, 'layers_pattern': None, 'rank_pattern': {}, 'alpha_pattern': {}, 'megatron_config': None, 'megatron_core': 'megatron.core', 'trainable_token_indices': None, 'loftq_config': {}, 'eva_config': None, 'corda_config': None, 'use_dora': False, 'alora_invocation_tokens': None, 'use_qalora': False, 'qalora_group_size': 16, 'layer_replication': None, 'runtime_config': {'ephemeral_gpu_offload': False}, 'lora_bias': False, 'target_parameters': None, 'arrow_config': None, 'ensure_weight_tying': False}}, 'image_token_index': 10, 'projector_hidden_act': 'gelu', 'vision_feature_layer': -1, 'vision_config': {'hidden_size': 1024, 'intermediate_size': 4096, 'num_hidden_layers': 24, 'num_attention_heads': 16, 'num_channels': 3, 'patch_size': 14, 'image_size': 1540, 'attention_dropout': 0.0, 'hidden_act': 'silu', 'head_dim': 64, 'initializer_range': 0.02, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': None, 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': None, 'eos_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': '', 'model_type': 'pixtral', 'output_attentions': False}, 'text_config': {'vocab_size': 131072, 'max_position_embeddings': 393216, 'hidden_size': 5120, 'intermediate_size': 32768, 'num_hidden_layers': 40, 'num_attention_heads': 32, 'sliding_window': None, 'head_dim': 128, 'num_key_value_heads': 8, 'hidden_act': 'silu', 'initializer_range': 0.02, 'rms_norm_eps': 1e-05, 'use_cache': True, 'attention_dropout': 0.0, 'rope_parameters': {'beta_fast': 32.0, 'beta_slow': 1.0, 'factor': 48.0, 'llama_4_scaling_beta': 0.1, 'mscale': 1.0, 'mscale_all_dim': 1.0, 'original_max_position_embeddings': 8192, 'rope_theta': 100000000.0, 'rope_type': 'yarn', 'type': 'yarn'}, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': False, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': None, 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 11, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': '', 'model_type': 'ministral3', 'output_attentions': False}, 'multimodal_projector_bias': False, 'spatial_merge_size': 2, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': False, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': ['Mistral3ForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': None, 'eos_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': 'Models/Devstral-Small-2-24B-HS-CPT', 'transformers_version': '5.0.0.dev0', 'model_type': 'mistral3', 'use_cache': False, 'output_attentions': False, 'output_dir': 'task2file/sft_devstral_24B_v2/checkpoints', 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 0.0001, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 0.8, 'num_train_epochs': 6.0, 'max_steps': -1, 'lr_scheduler_type': 'cosine', 'lr_scheduler_kwargs': None, 'warmup_ratio': 0.08, 'warmup_steps': 0.08, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': None, 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 2, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'enable_jit_checkpoint': False, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'use_cpu': False, 'seed': 42, 'data_seed': None, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': -1, 'ddp_backend': None, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 100, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'run_name': None, 'disable_tqdm': False, 'remove_unused_columns': False, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'eval_loss', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'project': 'huggingface', 'trackio_space_id': 'trackio', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'hub_revision': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'auto_find_batch_size': False, 'full_determinism': False, 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_num_input_tokens_seen': 'no', 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'liger_kernel_config': None, 'eval_use_gather_object': False, 'average_tokens_across_devices': True}
+2025-12-26 18:08:52,965 INFO    MainThread:190322 [wandb_config.py:__setitem__():154] [no run ID] config set model/num_parameters = 24022764544 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7b8940b75420>>
+2025-12-26 18:08:52,965 INFO    MainThread:190322 [wandb_run.py:_config_callback():1396] config_cb model/num_parameters 24022764544 None
+2025-12-28 04:02:04,643 INFO    MainThread:190322 [wandb_run.py:_finish():2287] finishing run sirajuddin-shaik-007/sft-training/ny9q48hd
+2025-12-28 04:02:04,645 INFO    MainThread:190322 [wandb_run.py:_atexit_cleanup():2486] got exitcode: 0
+2025-12-28 04:02:04,646 INFO    MainThread:190322 [wandb_run.py:_restore():2468] restore
+2025-12-28 04:02:04,646 INFO    MainThread:190322 [wandb_run.py:_restore():2474] restore done
+2025-12-28 04:02:05,050 INFO    MainThread:190322 [wandb_run.py:_footer_sync_info():3862] logging synced files