Training in progress, step 100

SFT LoRA checkpoint - step 100 (eval_loss: 1.7438)

Model save

Training in progress, step 2262

SFT LoRA checkpoint - step 2200 (eval_loss: 1.6671)

SFT LoRA checkpoint - step 2100 (eval_loss: 1.6668)

SFT LoRA checkpoint - step 2000 (eval_loss: 1.6671)

SFT LoRA checkpoint - step 1900 (eval_loss: 1.6668)

SFT LoRA checkpoint - step 1800 (eval_loss: 1.6673)

SFT LoRA checkpoint - step 1700 (eval_loss: 1.6675)

SFT LoRA checkpoint - step 1600 (eval_loss: 1.6669)

SFT LoRA checkpoint - step 1500 (eval_loss: 1.6682)

SFT LoRA checkpoint - step 1400 (eval_loss: 1.6684)

SFT LoRA checkpoint - step 1300 (eval_loss: 1.6703)

SFT LoRA checkpoint - step 1200 (eval_loss: 1.6724)

SFT LoRA checkpoint - step 1100 (eval_loss: 1.6705)

SFT LoRA checkpoint - step 1000 (eval_loss: 1.6729)

SFT LoRA checkpoint - step 900 (eval_loss: 1.6743)

SFT LoRA checkpoint - step 800 (eval_loss: 1.6776)

SFT LoRA checkpoint - step 700 (eval_loss: 1.6799)

SFT LoRA checkpoint - step 600 (eval_loss: 1.6824)

SFT LoRA checkpoint - step 500 (eval_loss: 1.6876)

SFT LoRA checkpoint - step 400 (eval_loss: 1.6937)

SFT LoRA checkpoint - step 300 (eval_loss: 1.7020)

SFT LoRA checkpoint - step 200 (eval_loss: 1.7135)

Training in progress, step 100

Tokenizer - SFT LoRA checkpoint - step 100 (eval_loss: 1.7438)

SFT LoRA checkpoint - step 100 (eval_loss: 1.7438)

initial commit