INDEX

Explanations

actions of resistance or defiance

oai_token-act-pair · gpt-3.5-turbo

references to law enforcement and situations involving authority and confrontation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yrinth

-0.75

ixture

-0.65

isible

-0.63

 Everywhere

-0.60

estinal

-0.60

 Header

-0.59

@#&

-0.59

cellent

-0.58

)--

-0.58

'>

-0.57

POSITIVE LOGITS

 himself

0.85

 anymore

0.84

his

0.65

 subordinates

0.64

 herself

0.63

 briefings

0.62

 superiors

0.62

 teammates

0.61

 resignation

0.61

 judgment

0.60

Activations Density 1.383%

actions of resistance or defiance

references to law enforcement and situations involving authority and confrontation

No Comments

No Known Activations