zephyr-7b-align-scan-0.0-0.0-linear-3
This model is a fine-tuned version of alignment-handbook/zephyr-7b-sft-full on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:
- Loss: 0.6455
- Rewards/chosen: -0.1452
- Rewards/rejected: -0.3169
- Rewards/accuracies: 0.3313
- Rewards/margins: 0.1718
- Logps/rejected: -103.6337
- Logps/chosen: -84.7995
- Logits/rejected: -2.2391
- Logits/chosen: -2.2616
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: 2.8387779271947853e-07
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.6466 | 1.0417 | 100 | 0.6659 | -0.0460 | -0.1148 | 0.3234 | 0.0688 | -89.2840 | -77.7589 | -2.4671 | -2.4857 |
| 0.5655 | 2.0833 | 200 | 0.6495 | -0.1460 | -0.2967 | 0.3294 | 0.1507 | -102.2009 | -84.8621 | -2.3195 | -2.3414 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0
- Datasets 2.21.0
- Tokenizers 0.19.1
- Downloads last month
- 2
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for taicheng/zephyr-7b-align-scan-0.0-0.0-linear-3
Base model
mistralai/Mistral-7B-v0.1
Finetuned
alignment-handbook/zephyr-7b-sft-full