-
Upload shared/model_config.yaml with huggingface_hub
-
Upload README.md with huggingface_hub
-
Delete train_pico.py with huggingface_hub
-
Delete tokenizer.json with huggingface_hub
-
Delete sft_fixed_step_9375.pt with huggingface_hub
-
Delete scripts/fix_and_retrain.py with huggingface_hub
-
Delete model.pt with huggingface_hub
-
Delete dpo_model.pt with huggingface_hub
-
Remove brand-name comparisons, describe directly
-
Reframed as edge personal assistant (Jarvis/Friday), not coding model
-
Update model card with latest training status and platform context
-
Add Pico architecture config
-
Add shared training configs and tokenizer
-
Upload scripts/fix_and_retrain.py with huggingface_hub
-
Upload sft_fixed_step_9375.pt with huggingface_hub
-
Add DPO checkpoint
-
Upload training script
-
Upload model checkpoint (293MB)
-
Upload training script
-
Add model card
-
Upload tokenizer
-
Upload model checkpoint (293MB)
-
initial commit
