Update README.md
Browse files
README.md
CHANGED
@@ -9,9 +9,9 @@ tags:
|
|
9 |
- Dolly
|
10 |
---
|
11 |
|
12 |
-
|
13 |
|
14 |
-
|
15 |
|
16 |
<pre>
|
17 |
!pip install -q -U bitsandbytes
|
@@ -20,7 +20,7 @@ tags:
|
|
20 |
!pip install -q -U git+https://github.com/huggingface/accelerate.git
|
21 |
</pre>
|
22 |
|
23 |
-
|
24 |
|
25 |
<pre>
|
26 |
<code>
|
@@ -39,7 +39,7 @@ model = PeftModel.from_pretrained(model, peft_model_id)
|
|
39 |
</pre>
|
40 |
|
41 |
|
42 |
-
|
43 |
|
44 |
<pre>
|
45 |
<code>
|
@@ -52,7 +52,7 @@ model = PeftModel.from_pretrained(model, peft_model_id)
|
|
52 |
</pre>
|
53 |
|
54 |
|
55 |
-
|
56 |
|
57 |
<pre>
|
58 |
<code>
|
|
|
9 |
- Dolly
|
10 |
---
|
11 |
|
12 |
+
**Requirements:**
|
13 |
|
14 |
+
You can run this model on Google Colab Pro, it requires a substantial amount of VRAM.
|
15 |
|
16 |
<pre>
|
17 |
!pip install -q -U bitsandbytes
|
|
|
20 |
!pip install -q -U git+https://github.com/huggingface/accelerate.git
|
21 |
</pre>
|
22 |
|
23 |
+
**Import this model using:**
|
24 |
|
25 |
<pre>
|
26 |
<code>
|
|
|
39 |
</pre>
|
40 |
|
41 |
|
42 |
+
**Inference using:**
|
43 |
|
44 |
<pre>
|
45 |
<code>
|
|
|
52 |
</pre>
|
53 |
|
54 |
|
55 |
+
**Output:**
|
56 |
|
57 |
<pre>
|
58 |
<code>
|