Commit
·
e4c9dfb
1
Parent(s):
eb4840a
Update README.md
Browse files
README.md
CHANGED
|
@@ -8,7 +8,9 @@ pipeline_tag: text-generation
|
|
| 8 |
|
| 9 |
## Model description
|
| 10 |
|
| 11 |
-
togethercomputer/RedPajama-INCITE-Base-3B-v1 finetuned for
|
|
|
|
|
|
|
| 12 |
|
| 13 |
Sample training data:
|
| 14 |
```json
|
|
@@ -49,7 +51,6 @@ Sample training data:
|
|
| 49 |
|
| 50 |
## Training procedure
|
| 51 |
|
| 52 |
-
|
| 53 |
The following `bitsandbytes` quantization config was used during training:
|
| 54 |
- load_in_8bit: False
|
| 55 |
- load_in_4bit: True
|
|
@@ -60,7 +61,7 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 60 |
- bnb_4bit_quant_type: nf4
|
| 61 |
- bnb_4bit_use_double_quant: True
|
| 62 |
- bnb_4bit_compute_dtype: bfloat16
|
| 63 |
-
### Framework versions
|
| 64 |
|
|
|
|
| 65 |
|
| 66 |
- PEFT 0.4.0.dev0
|
|
|
|
| 8 |
|
| 9 |
## Model description
|
| 10 |
|
| 11 |
+
The togethercomputer/RedPajama-INCITE-Base-3B-v1 model finetuned for `Paraphrasing` and `Changing the Tone` of the input sentence(to `casual`/`professional`/`witty`).
|
| 12 |
+
|
| 13 |
+
Look at the repo [llm-toys](https://github.com/kuutsav/llm-toys) for usage and other details.
|
| 14 |
|
| 15 |
Sample training data:
|
| 16 |
```json
|
|
|
|
| 51 |
|
| 52 |
## Training procedure
|
| 53 |
|
|
|
|
| 54 |
The following `bitsandbytes` quantization config was used during training:
|
| 55 |
- load_in_8bit: False
|
| 56 |
- load_in_4bit: True
|
|
|
|
| 61 |
- bnb_4bit_quant_type: nf4
|
| 62 |
- bnb_4bit_use_double_quant: True
|
| 63 |
- bnb_4bit_compute_dtype: bfloat16
|
|
|
|
| 64 |
|
| 65 |
+
### Framework versions
|
| 66 |
|
| 67 |
- PEFT 0.4.0.dev0
|