Commit
·
1f85620
1
Parent(s):
f6425c3
Update README.md
Browse files
README.md
CHANGED
@@ -3,6 +3,10 @@ datasets:
|
|
3 |
- argilla/ultrafeedback-binarized-preferences-cleaned
|
4 |
language:
|
5 |
- en
|
|
|
|
|
|
|
|
|
6 |
base_model: mistralai/Mixtral-8x7B-Instruct-v0.1
|
7 |
library_name: transformers
|
8 |
pipeline_tag: text-generation
|
@@ -34,10 +38,10 @@ This is part of the Notus family of models and experiments, where the Argilla te
|
|
34 |
|
35 |
### Model Description
|
36 |
|
37 |
-
- **Developed by:** Argilla (based on
|
38 |
- **Shared by:** Argilla
|
39 |
- **Model type:** Pretrained generative Sparse Mixture of Experts
|
40 |
-
- **Language(s) (NLP):**
|
41 |
- **License:** MIT
|
42 |
- **Finetuned from model:** [mistralai/Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1)
|
43 |
|
@@ -50,13 +54,12 @@ This is part of the Notus family of models and experiments, where the Argilla te
|
|
50 |
|
51 |
### Training Hardware
|
52 |
|
53 |
-
We used a VM with 8 x H100 80GB hosted in runpod.io for 1 epoch (~10hr)
|
54 |
|
55 |
### Training Data
|
56 |
|
57 |
We used a new iteration of the Argilla UltraFeedback preferences dataset named [argilla/ultrafeedback-binarized-preferences-cleaned](https://huggingface.co/datasets/argilla/ultrafeedback-binarized-preferences-cleaned).
|
58 |
|
59 |
-
|
60 |
## Training procedure
|
61 |
|
62 |
### Training hyperparameters
|
@@ -90,4 +93,4 @@ The following hyperparameters were used during training:
|
|
90 |
- Transformers 4.36.0
|
91 |
- Pytorch 2.1.0+cu118
|
92 |
- Datasets 2.14.6
|
93 |
-
- Tokenizers 0.15.0
|
|
|
3 |
- argilla/ultrafeedback-binarized-preferences-cleaned
|
4 |
language:
|
5 |
- en
|
6 |
+
- de
|
7 |
+
- es
|
8 |
+
- fr
|
9 |
+
- it
|
10 |
base_model: mistralai/Mixtral-8x7B-Instruct-v0.1
|
11 |
library_name: transformers
|
12 |
pipeline_tag: text-generation
|
|
|
38 |
|
39 |
### Model Description
|
40 |
|
41 |
+
- **Developed by:** Argilla (based on MistralAI previous efforts)
|
42 |
- **Shared by:** Argilla
|
43 |
- **Model type:** Pretrained generative Sparse Mixture of Experts
|
44 |
+
- **Language(s) (NLP):** English, Spanish, Italian, German, and French
|
45 |
- **License:** MIT
|
46 |
- **Finetuned from model:** [mistralai/Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1)
|
47 |
|
|
|
54 |
|
55 |
### Training Hardware
|
56 |
|
57 |
+
We used a VM with 8 x H100 80GB hosted in runpod.io for 1 epoch (~10hr).
|
58 |
|
59 |
### Training Data
|
60 |
|
61 |
We used a new iteration of the Argilla UltraFeedback preferences dataset named [argilla/ultrafeedback-binarized-preferences-cleaned](https://huggingface.co/datasets/argilla/ultrafeedback-binarized-preferences-cleaned).
|
62 |
|
|
|
63 |
## Training procedure
|
64 |
|
65 |
### Training hyperparameters
|
|
|
93 |
- Transformers 4.36.0
|
94 |
- Pytorch 2.1.0+cu118
|
95 |
- Datasets 2.14.6
|
96 |
+
- Tokenizers 0.15.0
|