Commit
•
77012a9
1
Parent(s):
a3c30bb
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,26 +1,41 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
-
base_model: BramVanroy/fietje-2b-
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
-
- trl
|
| 7 |
-
- dpo
|
| 8 |
-
- generated_from_trainer
|
| 9 |
- trl
|
|
|
|
|
|
|
| 10 |
- dpo
|
| 11 |
-
- generated_from_trainer
|
| 12 |
datasets:
|
| 13 |
- BramVanroy/ultra_feedback_dutch_cleaned
|
| 14 |
- BramVanroy/orca_dpo_pairs_dutch_cleaned
|
| 15 |
model-index:
|
| 16 |
-
- name: fietje-2b-
|
| 17 |
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
---
|
| 19 |
|
| 20 |
-
|
| 21 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
-
# fietje-2b-dpo-lr2.0e-6-beta0.2-gradaccum2-v6
|
| 24 |
|
| 25 |
This model is a fine-tuned version of [BramVanroy/fietje-2b-sft](https://huggingface.co/BramVanroy/fietje-2b-sft) on the BramVanroy/ultra_feedback_dutch_cleaned and the BramVanroy/orca_dpo_pairs_dutch_cleaned datasets.
|
| 26 |
It achieves the following results on the evaluation set:
|
|
@@ -75,4 +90,4 @@ The following hyperparameters were used during training:
|
|
| 75 |
- Transformers 4.39.1
|
| 76 |
- Pytorch 2.1.2+cu121
|
| 77 |
- Datasets 2.18.0
|
| 78 |
-
- Tokenizers 0.15.2
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
base_model: BramVanroy/fietje-2b-instruct
|
| 4 |
tags:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
- trl
|
| 6 |
+
- fietje
|
| 7 |
+
- alignment-handbook
|
| 8 |
- dpo
|
|
|
|
| 9 |
datasets:
|
| 10 |
- BramVanroy/ultra_feedback_dutch_cleaned
|
| 11 |
- BramVanroy/orca_dpo_pairs_dutch_cleaned
|
| 12 |
model-index:
|
| 13 |
+
- name: fietje-2b-chat
|
| 14 |
results: []
|
| 15 |
+
pipeline_tag: text-generation
|
| 16 |
+
inference: false
|
| 17 |
+
language:
|
| 18 |
+
- nl
|
| 19 |
---
|
| 20 |
|
| 21 |
+
<p align="center" style="margin:0;padding:0">
|
| 22 |
+
<img src="https://huggingface.co/BramVanroy/fietje-2b/resolve/main/img/fietje-2b-banner.png" alt="Fietje banner" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
|
| 23 |
+
</p>
|
| 24 |
+
|
| 25 |
+
<div style="margin:auto; text-align:center">
|
| 26 |
+
<h1 style="margin-bottom: 0">Fietje 2B Chat</h1>
|
| 27 |
+
<em>An open and efficient LLM for Dutch</em>
|
| 28 |
+
</div>
|
| 29 |
+
|
| 30 |
+
<blockquote class="tip">
|
| 31 |
+
<p align="center">
|
| 32 |
+
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b">👱♀️ Base version</a> -
|
| 33 |
+
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-instruct">🤖 Instruct version</a> -
|
| 34 |
+
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-chat">💬 Chat version</a> (this one) -
|
| 35 |
+
<a rel="nofollow" href="https://huggingface.co/BramVanroy/fietje-2b-chat-GGUF">🚀 GGUF of chat model</a>
|
| 36 |
+
</p>
|
| 37 |
+
</blockquote>
|
| 38 |
|
|
|
|
| 39 |
|
| 40 |
This model is a fine-tuned version of [BramVanroy/fietje-2b-sft](https://huggingface.co/BramVanroy/fietje-2b-sft) on the BramVanroy/ultra_feedback_dutch_cleaned and the BramVanroy/orca_dpo_pairs_dutch_cleaned datasets.
|
| 41 |
It achieves the following results on the evaluation set:
|
|
|
|
| 90 |
- Transformers 4.39.1
|
| 91 |
- Pytorch 2.1.2+cu121
|
| 92 |
- Datasets 2.18.0
|
| 93 |
+
- Tokenizers 0.15.2
|