Downloads · 30 days
10
24% of all-time downloads
voicebook-lora-v10-dpo
zehp/voicebook-lora-v10-dpo
voicebook-lora-v10-dpo is a machine learning model from zehp. 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.
DPO continue-training from voicebook-lora-v8 with proper reference policy.
All-time downloads
42
Public
Repo size
1 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1 GB · 98%
At a glance
- Library
- peft
- Access
- Public
- Created
- May 10, 2026
- Updated
- May 10, 2026
- SHA
- db92b11d
Base models
Model card
From the Hugging Face model README
voicebook-lora-v10-dpo
DPO continue-training from voicebook-lora-v8 with proper reference policy.
Fixes 4 mistakes from v9-DPO:
- Dataset: 1399 pairs (vs 189) — FAIL + WARN + synthetic-broken-PASS
- Continue-from-v8: v8 LoRA loaded as TRAINABLE adapter (not merged-and-wiped), reference logprobs precomputed with v8 ENABLED
- Schedule: 3 epochs, lr=2e-5, cosine warmup_ratio=0.1
- eval_strategy=no (custom trainer eval was buggy; train metrics monitored)
Final metrics:
- train_loss: 0.0607
- rewards/accuracies (final): 1.0
- rewards/margins (final): ~14-16
- Library
- peft
- Created
- May 10, 2026
- Updated
- May 10, 2026