INDEX

Explanations

phrases related to employing or making use of skills or resources

oai_token-act-pair · gpt-3.5-turbo

phrases indicating actions or putting something to use

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

oslov

-0.61

artifacts

-0.58

SourceFile

-0.57

 sshd

-0.56

 Polic

-0.55

 Theft

-0.55

 Latest

-0.54

initialized

-0.53

jc

-0.53

grain

-0.52

POSITIVE LOGITS

asting

0.80

 shame

0.80

bed

0.76

perty

0.75

 rout

0.75

asty

0.75

 pasture

0.74

oult

0.74

tle

0.74

asted

0.72

Activations Density 0.059%

phrases related to employing or making use of skills or resources

phrases indicating actions or putting something to use

No Known Activations