-
Update README.md
-
Upload eval_results.json with huggingface_hub (#1)
- Upload eval_results.json with huggingface_hub (c5eb9b70f0ba0abc0b69fbf90a1b362731abeb99) -
Upload logs
-
Upload logs
-
step 5764 | val_ppl=11.03
-
step 5600 | val_ppl=11.05
-
step 5200 | val_ppl=11.10
-
step 4800 | val_ppl=11.18
-
step 4400 | val_ppl=11.29
-
step 4000 | val_ppl=11.43
-
step 3600 | val_ppl=11.63
-
step 3200 | val_ppl=11.88
-
step 2800 | val_ppl=12.23
-
step 2400 | val_ppl=12.71
-
step 2000 | val_ppl=13.38
-
step 1600 | val_ppl=14.38
-
step 1200 | val_ppl=15.94
-
step 800 | val_ppl=18.87
-
step 400 | val_ppl=32.14
-
Delete training_meta.json
-
Delete tokenizer_config.json
-
Delete tokenizer.json
-
Delete model.safetensors
-
Delete generation_config.json
-
Delete config.json
-
Delete resume
-
step 20 | val_ppl=inf
-
Delete tokenizer_config.json
-
Delete tokenizer.json
-
Delete model.safetensors
-
Delete generation_config.json
-
Delete model.py
-
Delete config.json
-
Delete resume
-
Delete training_meta.json
-
Update model.py
-
step 20 | ppl=inf
-
step 10 | ppl=inf
-
initial commit
