Downloads · 30 days
17
29% of all-time downloads
Meta-Llama-3.1-70B-FLUTE
radi-cho/Meta-Llama-3.1-70B-FLUTE
Meta-Llama-3.1-70B-FLUTE is a machine learning model from radi-cho. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as llama3.1.
Revisions available in this repository: - main (W4G64, scales learned); - nflw3g64 (W3G64, scales learned);
All-time downloads
59
Public
Parameters
20.3B
72.6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors40.6 GB · 100%
Parameter types
How the weights are stored.
I1617.1B · 84%
Model card
From the Hugging Face model README
| Wiki | C4 | PIQA | ARC-E | ARC-C | HellaSwag | Wino | Avg. | |
|---|---|---|---|---|---|---|---|---|
| Unquantized | 2.82 | 7.18 | 82.81 | 85.31 | 59.64 | 67.49 | 82.00 | 75.45 |
| W4G64 | 3.09 | 7.53 | 83.03 | 85.52 | 58.19 | 67.04 | 80.43 | 74.84 |
| W3G64 | 4.29 | 8.91 | 82.04 | 83.29 | 54.78 | 64.99 | 78.14 | 72.65 |
Revisions available in this repository:
main(W4G64, scales learned);nfl_w3g64(W3G64, scales learned);
Evaluations are provided for models with learned scales.<br>Benchmark scores (zero-shot) are computed with lm-evaluation-harness.
- Type
- llama
- License
- llama3.1
- Created
- Aug 4, 2024
- Updated
- Aug 5, 2024