Model save

README.md

@@ -0,0 +1,90 @@
+---
+library_name: peft
+base_model: peiyi9979/math-shepherd-mistral-7b-prm
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+- precision
+- recall
+- f1
+model-index:
+- name: v1_5_mistral_lora
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# v1_5_mistral_lora
+
+This model is a fine-tuned version of [peiyi9979/math-shepherd-mistral-7b-prm](https://huggingface.co/peiyi9979/math-shepherd-mistral-7b-prm) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3143
+- Accuracy: 0.8639
+- Precision: 0.7383
+- Recall: 0.7453
+- F1: 0.7418
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 765837
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 32
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
+| No log        | 0      | 0    | 0.5958          | 0.7376   | 0.5       | 0.0660 | 0.1167 |
+| 0.6808        | 0.0575 | 20   | 0.5704          | 0.7401   | 0.5385    | 0.0660 | 0.1176 |
+| 0.4764        | 0.1149 | 40   | 0.4768          | 0.7574   | 0.5303    | 0.6604 | 0.5882 |
+| 0.4099        | 0.1724 | 60   | 0.4052          | 0.8020   | 0.6275    | 0.6038 | 0.6154 |
+| 0.346         | 0.2299 | 80   | 0.3761          | 0.8366   | 0.6961    | 0.6698 | 0.6827 |
+| 0.2929        | 0.2874 | 100  | 0.3664          | 0.8366   | 0.6887    | 0.6887 | 0.6887 |
+| 0.3822        | 0.3448 | 120  | 0.3551          | 0.8515   | 0.6983    | 0.7642 | 0.7297 |
+| 0.3955        | 0.4023 | 140  | 0.3442          | 0.8589   | 0.7634    | 0.6698 | 0.7136 |
+| 0.34          | 0.4598 | 160  | 0.3399          | 0.8614   | 0.7232    | 0.7642 | 0.7431 |
+| 0.2897        | 0.5172 | 180  | 0.3244          | 0.8614   | 0.7404    | 0.7264 | 0.7333 |
+| 0.2599        | 0.5747 | 200  | 0.3225          | 0.8639   | 0.7383    | 0.7453 | 0.7418 |
+| 0.34          | 0.6322 | 220  | 0.3178          | 0.8688   | 0.7573    | 0.7358 | 0.7464 |
+| 0.2969        | 0.6897 | 240  | 0.3178          | 0.8564   | 0.7222    | 0.7358 | 0.7290 |
+| 0.3179        | 0.7471 | 260  | 0.3128          | 0.8663   | 0.7453    | 0.7453 | 0.7453 |
+| 0.2901        | 0.8046 | 280  | 0.3146          | 0.8639   | 0.7383    | 0.7453 | 0.7418 |
+| 0.2587        | 0.8621 | 300  | 0.3138          | 0.8639   | 0.7383    | 0.7453 | 0.7418 |
+| 0.326         | 0.9195 | 320  | 0.3151          | 0.8589   | 0.7248    | 0.7453 | 0.7349 |
+| 0.3475        | 0.9770 | 340  | 0.3143          | 0.8639   | 0.7383    | 0.7453 | 0.7418 |
+
+
+### Framework versions
+
+- PEFT 0.13.2
+- Transformers 4.46.0
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

adapter_config.json

@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "peiyi9979/math-shepherd-mistral-7b-prm",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eaab67199443da4c430ad7f60281d599dea28e4b85d573f65f525c9fc9ea652
+size 27280152