Orca-2-13b-abliterated

noneUsername/Orca-2-13b-abliterated

Orca-2-13b-abliterated is a machine learning model from noneUsername. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

Downloads · 30 days

10% of all-time downloads

All-time downloads

Public

Parameters

13B

26 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors26 GB · 100%

Parameter types

How the weights are stored.

F169.2B · 71%

Base models

uukuguy/Orca-2-13b-f16

Model card

From the Hugging Face model README

multstage abliteration

vllm (pretrained=/mnt/g/abliterated/Orca-2-13b-f16,add_bos_token=true,max_model_len=2048,dtype=float16,tensor_parallel_size=2,gpu_memory_utilization=0.8,max_num_seqs=3), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: 1

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.496	±	0.0317
		strict-match	5	exact_match	↑	0.488	±	0.0317

vllm (pretrained=/mnt/g/abliterated/Orca-2-13b-abliterated-round1,add_bos_token=true,max_model_len=2048,dtype=float16,tensor_parallel_size=2,gpu_memory_utilization=0.8,max_num_seqs=3), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: 2

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.536	±	0.0316
		strict-match	5	exact_match	↑	0.528	±	0.0316

vllm (pretrained=/mnt/g/abliterated/Orca-2-13b-abliterated-round2,add_bos_token=true,max_model_len=2048,dtype=float16,tensor_parallel_size=2,gpu_memory_utilization=0.85,max_num_seqs=4), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: 4

Tasks	Version	Filter	n-shot	Metric		Value		Stderr
gsm8k	3	flexible-extract	5	exact_match	↑	0.512	±	0.0317
		strict-match	5	exact_match	↑	0.500	±	0.0317

Type: llama
Created: Mar 18, 2025
Updated: Mar 18, 2025