willyninja30
commited on
Commit
•
5067875
1
Parent(s):
c1d170d
Update README.md
Browse files
README.md
CHANGED
@@ -27,8 +27,7 @@ Llama 2 is a collection of pretrained and fine-tuned generative text models rang
|
|
27 |
# *FINETUNING PROCESS **
|
28 |
|
29 |
We trained the model on a high quality dataset with more than 50.000 rows of french language. The training took 2 days on Amazon Cloud Sagemaker powered by Nvidia GPUs.
|
30 |
-
|
31 |
-
....
|
32 |
# **Timing of training**
|
33 |
2 Days using NVIDIA A10G and Amazon Web services Cloud Instance. We are grateful to Nvidia Inception program.
|
34 |
|
@@ -37,19 +36,19 @@ We are also applying rope scalling as experimental approach used by several othe
|
|
37 |
*Note: Use of this model is governed by the Meta license because it's based on LLAMA 2. In order to download the model weights and tokenizer, please visit the [website](https://ai.meta.com/resources/models-and-libraries/llama-downloads/) and accept our License before requesting access here.*
|
38 |
|
39 |
|
40 |
-
**Model Developers** FARADAY
|
41 |
|
42 |
-
**Variations**
|
43 |
|
44 |
-
**Input** Models input text only.
|
45 |
|
46 |
-
**Output** Models generate text only.
|
47 |
|
48 |
-
**Model Architecture** ARIA is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.
|
49 |
|
50 |
-
**License** A custom commercial license is available at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
|
51 |
|
52 |
-
**Research Paper for LLAMA 2** ["Llama-2: Open Foundation and Fine-tuned Chat Models"](arxiv.org/abs/2307.09288)
|
53 |
|
54 |
|
55 |
|
|
|
27 |
# *FINETUNING PROCESS **
|
28 |
|
29 |
We trained the model on a high quality dataset with more than 50.000 rows of french language. The training took 2 days on Amazon Cloud Sagemaker powered by Nvidia GPUs.
|
30 |
+
|
|
|
31 |
# **Timing of training**
|
32 |
2 Days using NVIDIA A10G and Amazon Web services Cloud Instance. We are grateful to Nvidia Inception program.
|
33 |
|
|
|
36 |
*Note: Use of this model is governed by the Meta license because it's based on LLAMA 2. In order to download the model weights and tokenizer, please visit the [website](https://ai.meta.com/resources/models-and-libraries/llama-downloads/) and accept our License before requesting access here.*
|
37 |
|
38 |
|
39 |
+
**Model Developers** :FARADAY
|
40 |
|
41 |
+
**Variations** :ARIA comes in a range of parameter sizes — 7B, 40B (based on Falcon), and 70B finetuned on French language datasets.
|
42 |
|
43 |
+
**Input** :Models input text only.
|
44 |
|
45 |
+
**Output** : Models generate text only.
|
46 |
|
47 |
+
**Model Architecture** : ARIA is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.
|
48 |
|
49 |
+
**License** : A custom commercial license is available at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
|
50 |
|
51 |
+
**Research Paper for LLAMA 2** : ["Llama-2: Open Foundation and Fine-tuned Chat Models"](arxiv.org/abs/2307.09288)
|
52 |
|
53 |
|
54 |
|