Commit
•
473a2ee
1
Parent(s):
60dde8d
Update README.md (#1)
Browse files- Update README.md (8af75876bfdfc33c4f81882cdd7d464018154f2c)
Co-authored-by: Samuele Colombo <FinancialSupport@users.noreply.huggingface.co>
README.md
CHANGED
@@ -12,6 +12,9 @@ tags:
|
|
12 |
base_model: unsloth/llama-3-8b-bnb-4bit
|
13 |
datasets:
|
14 |
- mii-community/ultrafeedback-translated-ita
|
|
|
|
|
|
|
15 |
---
|
16 |
|
17 |
# Uploaded model
|
@@ -22,4 +25,11 @@ datasets:
|
|
22 |
|
23 |
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
25 |
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
|
|
12 |
base_model: unsloth/llama-3-8b-bnb-4bit
|
13 |
datasets:
|
14 |
- mii-community/ultrafeedback-translated-ita
|
15 |
+
|
16 |
+
|
17 |
+
## Evaluation
|
18 |
---
|
19 |
|
20 |
# Uploaded model
|
|
|
25 |
|
26 |
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
27 |
|
28 |
+
For a detailed comparison of model performance, check out the [Leaderboard for Italian Language Models](https://huggingface.co/spaces/FinancialSupport/open_ita_llm_leaderboard).
|
29 |
+
|
30 |
+
Here's a breakdown of the performance metrics:
|
31 |
+
| Metric | hellaswag_it acc_norm | arc_it acc_norm | m_mmlu_it 5-shot acc | Average |
|
32 |
+
|:----------------------------|:----------------------|:----------------|:---------------------|:--------|
|
33 |
+
| **Accuracy Normalized** | 0.6064 | 0.4611 | 0.5328 | 0.5334 |
|
34 |
+
|
35 |
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|