Info

BLOOMZ-3b fine-tuned on legal language. Experimental purposes only! This fine-tuning makes no attempt to understand the true nature of the dataset. It simply attempts to assess the performance of basic prompt engineering on BLOOMZ-3b for legal language.

Example Output

INSTRUCTION

Below are two legal citations. Please generate a summary judgment.

Citation A:

United States v. Mitchell, 502 F.3d 931, 991 (9th Cir.2007) (approving of district court not allowing witnesses to offer opinions on what jury’s verdict should be); United States v. Caro, 461 F.Supp.2d. 459, 465 (W.D.Va.2006) (“[A]n express plea for mercy to the jury from a defendant’s witness is not mitigating evidence that could aid the jury in their decision making.”); see also Stenson v. Lambert, 504 F.3d 873, 892 (9th Cir.2007) (noting there are no federal cases requiring the admission of execution impact testimony); Jackson v. Dretke, 450 F.3d 614, 617-18 (5th Cir.2006); but see Jackson, 450 F.3d at 620 (Dennis, J., dissenting).

Citation B:

United States v. Mitchell, 502 F.3d 931, 991 (9th Cir.2007) (approving of district court not allowing witnesses to offer opinions on what jury’s verdict should be); United States v. Caro, 461 F.Supp.2d. 459, 465 (W.D.Va.2006) (“[A]n express plea for mercy to the jury from a defendant’s witness is not mitigating evidence that could aid the jury in their decision making.”); see also Stenson v. Lambert, 504 F.3d 873, 892 (9th Cir.2007) (noting there are no federal cases requiring the admission of execution impact testimony); Jackson v. Dretke, 450 F.3d 614, 617-18 (5th Cir.2006); but see Jackson, 450 F.3d at 620 (Dennis, J., dissenting).

Judgment:

The district court's decision to allow the testimony of the defendant's execution impact witness is not mitigating evidence. (Model repeats for as many tokens as possible.)

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.

Dataset used to train amqdn/bloomz-legalsupport-lm-lora