Update README.md
Browse files
README.md
CHANGED
|
@@ -11,3 +11,33 @@ tags:
|
|
| 11 |
base_model: unsloth/gemma-7b-bnb-4bit
|
| 12 |
---
|
| 13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
base_model: unsloth/gemma-7b-bnb-4bit
|
| 12 |
---
|
| 13 |
|
| 14 |
+
|
| 15 |
+
```python
|
| 16 |
+
from unsloth import FastLanguageModel
|
| 17 |
+
import torch
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
model, tokenizer = FastLanguageModel.from_pretrained(
|
| 21 |
+
model_name = "Ashishkr/Gemma-7B-COT-LoRA-1024R",
|
| 22 |
+
max_seq_length = max_seq_length,
|
| 23 |
+
dtype = dtype,
|
| 24 |
+
load_in_4bit = load_in_4bit,
|
| 25 |
+
)
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
|
| 29 |
+
FastLanguageModel.for_inference(model)
|
| 30 |
+
inputs = tokenizer(
|
| 31 |
+
[
|
| 32 |
+
alpaca_prompt.format(
|
| 33 |
+
"given a number identify if it is an even number, if its even, square it and add 5 to it. if its odd , get me the square root of the number", # instruction
|
| 34 |
+
"12", # input
|
| 35 |
+
"", # output - leave this blank for generation!
|
| 36 |
+
)
|
| 37 |
+
], return_tensors = "pt").to("cuda")
|
| 38 |
+
|
| 39 |
+
outputs = model.generate(**inputs, max_new_tokens = 2048, use_cache = True)
|
| 40 |
+
tokenizer.batch_decode(outputs)
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
```
|