smcleish's picture
Update README.md
70c0c5a verified
---
library_name: transformers
license: mit
base_model:
- meta-llama/Meta-Llama-3-8B
---
# Model Details
`meta-llama/Meta-Llama-3-8B` model finetuned on 100,000 [CLRS-Text](https://github.com/google-deepmind/clrs/tree/master/clrs/_src/clrs_text) examples.
## Training Details
- Learning Rate: 1e-4, 150 warmup steps then cosine decayed to 5e-06 using AdamW optimiser
- Batch size: 128
- Loss taken over answer only, not on question.