35 ms·

It says "wait" (as in "wait, no, I should do X") so much while reasoning it's almost comical. I also ran into the "catastrophic forgetting" issue that others h

It says "wait" (as in "wait, no, I should do X") so much while reasoning it's almost comical. I also ran into the "catastrophic forgetting" issue that others have reported - it sometimes loses the plot after producing a lot of reasoning tokens. Overall though quite impressive if you're not in a hurry.

rahimnathwani 2y ago
Is the model using budget forcing?
- rosspackard 2y ago
  I have a suspicion it does use budget forcing. The word "alternatively" also frequently show up and it happens when it seems logically that a </think> tag could have been place.
- Szpadel 2y ago
  I do not understand why to force wait when model want to output </think>. why not just decrease </think> probability? if model really wants to finish maybe or could over power it in cases were it's really simple question. and definitely would allow model to express next thought more freely
  - rahimnathwani 2y ago
    why not just decrease </think> probability? Huggingface's transformers library supports something similar to this. You set a minimum length, and until that length is reached, the end of sequence token has no chance of being output. https://github.com/huggingface/transformers/blob/51ed61e2f05176f81fa7c9decba10cc28e138f61/src/transformers/generation/logits_process.py#L95 https://github.com/huggingface/transformers/blob/51ed61e2f05... S1 does something similar to put a lower limit on its reasoning output. End of thinking is represented with the <|im_start|> token, followed by the word 'answer'. IIRC the code dynamically adds/removes <|im_start|> to the list of suppressed tokens. Both of these approaches set the probability to zero, not something small like you were suggesting.
huseyinkeles 2y ago
I read somewhere which I can't find now, that for the -reasoning- models they trained heavily to keep saying "wait" so they can keep reasoning and not return early.

rahimnathwani 2y ago

Is the model using budget forcing?

rosspackard 2y ago
I have a suspicion it does use budget forcing. The word "alternatively" also frequently show up and it happens when it seems logically that a </think> tag could have been place.
Szpadel 2y ago
I do not understand why to force wait when model want to output </think>. why not just decrease </think> probability? if model really wants to finish maybe or could over power it in cases were it's really simple question. and definitely would allow model to express next thought more freely
- rahimnathwani 2y ago
  why not just decrease </think> probability? Huggingface's transformers library supports something similar to this. You set a minimum length, and until that length is reached, the end of sequence token has no chance of being output. https://github.com/huggingface/transformers/blob/51ed61e2f05176f81fa7c9decba10cc28e138f61/src/transformers/generation/logits_process.py#L95 https://github.com/huggingface/transformers/blob/51ed61e2f05... S1 does something similar to put a lower limit on its reasoning output. End of thinking is represented with the <|im_start|> token, followed by the word 'answer'. IIRC the code dynamically adds/removes <|im_start|> to the list of suppressed tokens. Both of these approaches set the probability to zero, not something small like you were suggesting.

huseyinkeles 2y ago

I read somewhere which I can't find now, that for the -reasoning- models they trained heavily to keep saying "wait" so they can keep reasoning and not return early.