Update README.md
Browse files
README.md
CHANGED
@@ -20,7 +20,25 @@ It achieves the following results on the evaluation set:
|
|
20 |
|
21 |
## Model description
|
22 |
|
23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
|
25 |
## Intended uses & limitations
|
26 |
|
|
|
20 |
|
21 |
## Model description
|
22 |
|
23 |
+
Trained on 1K dataset instead of 400
|
24 |
+
|
25 |
+
lora_alpha = 256
|
26 |
+
|
27 |
+
lora_dropout = 0.05
|
28 |
+
|
29 |
+
lora_r = 128
|
30 |
+
|
31 |
+
epochs = 3
|
32 |
+
|
33 |
+
learning rate = 3e-4
|
34 |
+
|
35 |
+
warmup_steps=100
|
36 |
+
|
37 |
+
gradient_accumulation_steps = 1
|
38 |
+
|
39 |
+
train_batch = 2
|
40 |
+
|
41 |
+
eval_batch = 2
|
42 |
|
43 |
## Intended uses & limitations
|
44 |
|