Hugging Face
Models
Datasets
Spaces
Community
Docs
Enterprise
Pricing
Log In
Sign Up
OpenTransformer
/
AGILLM-3-large
like
0
Model card
Files
Files and versions
xet
Community
2b0bfd4
AGILLM-3-large
58.7 GB
1 contributor
History:
28 commits
OpenTransformer
Add experiments/n_heavy.py
2b0bfd4
verified
about 2 months ago
checkpoints
Upload checkpoints/pretrain_step01620126.pt with huggingface_hub
about 2 months ago
experiments
Add experiments/n_heavy.py
about 2 months ago
scripts
Backup script hf_upload.py
about 2 months ago
tokenizer
Add tokenizer: tokenizer.json
about 2 months ago
.gitattributes
Safe
1.52 kB
initial commit
about 2 months ago
CHANGELOG.md
Safe
612 Bytes
Add CHANGELOG documenting GradScaler fix
about 2 months ago
README.md
Safe
1.48 kB
Add README with model details
about 2 months ago
hf_upload.py
Safe
1.92 kB
Add checkpoint uploader script
about 2 months ago
hot_config.json
Safe
253 Bytes
Add hot-reload config template
about 2 months ago
n.py
Safe
46.9 kB
Fix GradScaler resume bug - wrapped scaler.load_state_dict() in try/except at line 512. Allows resuming from checkpoints saved without AMP.
about 2 months ago
rotating_log.py
Safe
2.9 kB
Add dual rotating log (5000+2500 lines)
about 2 months ago
