Update README.md
Browse files
README.md
CHANGED
|
@@ -6,6 +6,10 @@ datasets:
|
|
| 6 |
- togethercomputer/RedPajama-Data-1T
|
| 7 |
---
|
| 8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
# OpenLLaMA: An Open Reproduction of LLaMA
|
| 10 |
|
| 11 |
**TL;DR**: we are releasing our public preview of OpenLLaMA, a permissively licensed open source reproduction of Meta AI’s LLaMA. We are releasing a series of 3B, 7B and 13B models trained on different data mixtures. Our model weights can serve as the drop in replacement of LLaMA in existing implementations.
|
|
@@ -24,14 +28,7 @@ Preview checkpoints can be directly loaded from Hugging Face Hub. **Please note
|
|
| 24 |
import torch
|
| 25 |
from transformers import LlamaTokenizer, LlamaForCausalLM
|
| 26 |
|
| 27 |
-
|
| 28 |
-
model_path = 'openlm-research/open_llama_3b_v2'
|
| 29 |
-
# model_path = 'openlm-research/open_llama_7b_v2'
|
| 30 |
-
|
| 31 |
-
## v1 models
|
| 32 |
-
# model_path = 'openlm-research/open_llama_3b'
|
| 33 |
-
# model_path = 'openlm-research/open_llama_7b'
|
| 34 |
-
# model_path = 'openlm-research/open_llama_13b'
|
| 35 |
|
| 36 |
tokenizer = LlamaTokenizer.from_pretrained(model_path)
|
| 37 |
model = LlamaForCausalLM.from_pretrained(
|
|
|
|
| 6 |
- togethercomputer/RedPajama-Data-1T
|
| 7 |
---
|
| 8 |
|
| 9 |
+
# Tokenizer Fixed!! 🎉
|
| 10 |
+
Thanks to https://huggingface.co/mistralai/Mistral-7B-v0.1/discussions/26/files
|
| 11 |
+
|
| 12 |
+
|
| 13 |
# OpenLLaMA: An Open Reproduction of LLaMA
|
| 14 |
|
| 15 |
**TL;DR**: we are releasing our public preview of OpenLLaMA, a permissively licensed open source reproduction of Meta AI’s LLaMA. We are releasing a series of 3B, 7B and 13B models trained on different data mixtures. Our model weights can serve as the drop in replacement of LLaMA in existing implementations.
|
|
|
|
| 28 |
import torch
|
| 29 |
from transformers import LlamaTokenizer, LlamaForCausalLM
|
| 30 |
|
| 31 |
+
model_path = 'typeof/open_llama_3b_v2'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
|
| 33 |
tokenizer = LlamaTokenizer.from_pretrained(model_path)
|
| 34 |
model = LlamaForCausalLM.from_pretrained(
|