Trained at 8K. 32,768,000 tokens seen (7,871,550 of them unmasked).
Full model: xzuyn/Gemma-3-1B-Instruct-TrashMix-v1.1
Note: I accidentally used ChatML role names, so try <start_of_turn>assistant too.
- Downloads last month
- 4
Model tree for PJMixers-Dev/Gemma-3-1B-Instruct-TrashMix-v1.1-LoRA-run3
Base model
google/gemma-3-1b-pt
Finetuned
google/gemma-3-1b-it
Finetuned
PJMixers-Dev/gemma-3-1b-it-fixed