add handling for argilla dpo-mix (#1397)
Browse files
src/axolotl/prompt_strategies/dpo/chatml.py
CHANGED
|
@@ -24,6 +24,25 @@ def argilla(
|
|
| 24 |
return transform_fn
|
| 25 |
|
| 26 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
def icr(
|
| 28 |
cfg,
|
| 29 |
**kwargs,
|
|
|
|
| 24 |
return transform_fn
|
| 25 |
|
| 26 |
|
| 27 |
+
def argilla_chat(
|
| 28 |
+
cfg,
|
| 29 |
+
**kwargs,
|
| 30 |
+
): # pylint: disable=possibly-unused-variable,unused-argument
|
| 31 |
+
"""
|
| 32 |
+
for argilla/dpo-mix-7k conversations
|
| 33 |
+
"""
|
| 34 |
+
|
| 35 |
+
def transform_fn(sample):
|
| 36 |
+
sample[
|
| 37 |
+
"prompt"
|
| 38 |
+
] = f"<|im_start|>user\n{sample['chosen'][0]['content']}<|im_end|>\n<|im_start|>assistant\n"
|
| 39 |
+
sample["chosen"] = f"{sample['chosen'][1]['content']}<|im_end|>"
|
| 40 |
+
sample["rejected"] = f"{sample['rejected'][1]['content']}<|im_end|>"
|
| 41 |
+
return sample
|
| 42 |
+
|
| 43 |
+
return transform_fn
|
| 44 |
+
|
| 45 |
+
|
| 46 |
def icr(
|
| 47 |
cfg,
|
| 48 |
**kwargs,
|