Downloads · 30 days
2
7% of all-time downloads
npov_RM_google_S130104_LLM_false_STRUCT_false_epo3_lr5e-4_r16_2602201450
leobianco/npov_RM_google_S130104_LLM_false_STRUCT_false_epo3_lr5e-4_r16_2602201450
npov_RM_google_S130104_LLM_false_STRUCT_false_epo3_lr5e-4_r16_2602201450 is a machine learning model from leobianco. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as gemma.
should probably proofread and complete it, then remove this comment. --
All-time downloads
28
Public
Repo size
57.9 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.json34.4 MB · 76%
At a glance
- Library
- peft
- License
- gemma
- Access
- Public
- Created
- Feb 20, 2026
- Updated
- Feb 20, 2026
- SHA
- 85fcea8c
Base models
Model card
From the Hugging Face model README
<!-- 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. -->
npov_RM_google_S130104_LLM_false_STRUCT_false_epo3_lr5e-4_r16_2602201450
This model is a fine-tuned version of google/gemma-2-2b-it on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.2880
- Roc Auc: 0.8103
- Best Threshold: 0.4636
- Tpr At Best Threshold: 0.9573
- Fpr At Best Threshold: 0.4667
- Accuracy At Best Threshold: 0.8707
- Avg Score True Positives: 0.7880
- Avg Score True Negatives: 0.5351
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: 0.0005
- train_batch_size: 4
- eval_batch_size: 4
- seed: 130104
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Roc Auc | Best Threshold | Tpr At Best Threshold | Fpr At Best Threshold | Accuracy At Best Threshold | Avg Score True Positives | Avg Score True Negatives |
|---|---|---|---|---|---|---|---|---|---|---|
| No log | 0 | 0 | 5.5312 | 0.5 | inf | 0.0 | 0.0 | 0.2041 | 0.5 | 0.5 |
| 4.7217 | 0.3571 | 5 | 4.5717 | 0.6125 | 0.9073 | 0.8205 | 0.4667 | 0.7619 | 0.9142 | 0.9058 |
| 4.1777 | 0.7143 | 10 | 3.9735 | 0.6986 | 0.9005 | 0.4274 | 0.1333 | 0.5170 | 0.8853 | 0.8406 |
| 2.8286 | 1.0714 | 15 | 4.1424 | 0.5201 | 0.7950 | 0.8547 | 0.6333 | 0.7551 | 0.8137 | 0.8123 |
| 4.0322 | 1.4286 | 20 | 4.0565 | 0.5788 | 0.7780 | 0.7009 | 0.4667 | 0.6667 | 0.7892 | 0.7791 |
| 3.3979 | 1.7857 | 25 | 3.6859 | 0.7491 | 0.8093 | 0.8034 | 0.3333 | 0.7755 | 0.8434 | 0.7702 |
| 2.5884 | 2.1429 | 30 | 3.4106 | 0.7887 | 0.9086 | 0.6068 | 0.1333 | 0.6599 | 0.8851 | 0.7366 |
| 2.2051 | 2.5 | 35 | 3.1335 | 0.8117 | 0.6406 | 0.9231 | 0.4333 | 0.8503 | 0.8513 | 0.6202 |
| 3.2937 | 2.8571 | 40 | 3.2880 | 0.8103 | 0.4636 | 0.9573 | 0.4667 | 0.8707 | 0.7880 | 0.5351 |
Framework versions
- PEFT 0.18.1
- Transformers 4.57.6
- Pytorch 2.9.1+cu128
- Datasets 4.5.0
- Tokenizers 0.22.2
- Library
- peft
- License
- gemma
- Created
- Feb 20, 2026
- Updated
- Feb 20, 2026