Insights DB

Last scanned: Jun 23, 2025 at 8:46 PM UTC

Version	Commit message	Commit author	Last scanned
8f32c9	Final tokenizer (Enhanced) (Trained with Unsloth)	Gege24	Jun 23, 2025
df836e	Training completed - Final tokenizer	Gege24	Jun 23, 2025
e1137b	Training completed at step 900 - Final enhanced full model	Gege24	Jun 23, 2025
098127	Training in progress, step 900, checkpoint	Gege24	Jun 23, 2025
c653b1	Training in progress, step 900	Gege24	Jun 23, 2025
2404a0	NOM (Trained with Unsloth)	Gege24	Jun 23, 2025
08cf32	NOM (Trained with Unsloth)	Gege24	Jun 23, 2025
bf2487	NOM (Trained with Unsloth)	Gege24	Jun 23, 2025
85e934	NOM (Trained with Unsloth)	Gege24	Jun 23, 2025
e1b089	NOM (Trained with Unsloth)	Gege24	Never scanned
bd6bfd	NOM (Trained with Unsloth)	Gege24	Never scanned
5bdece	Checkpoint 500 - Tokenizer	Gege24	Never scanned
544c0f	Checkpoint 500 - Full model with merged adapters (Enhanced)	Gege24	Jun 23, 2025
3757c3	Training in progress, step 500, checkpoint	Gege24	Jun 23, 2025
c2df0a	Training in progress, step 500	Gege24	Never scanned
de1342	NOM (Trained with Unsloth)	Gege24	Never scanned
eccf64	NOM (Trained with Unsloth)	Gege24	Jun 23, 2025
f253a3	NOM (Trained with Unsloth)	Gege24	Jun 23, 2025
1842cd	NOM (Trained with Unsloth)	Gege24	Jun 23, 2025
24dabd	NOM (Trained with Unsloth)	Gege24	Never scanned

‌
‌
‌
‌
‌
‌
‌
‌
‌
‌

With Prisma AIRS you can scan models for threats before ML developers download them for use, and apply policies based on your risk tolerance.

Earn bounties and get recognition for your discoveries by submitting them through Huntr.

Products and Services