INDEX

Explanations

words related to negative judgment of intelligence or actions

oai_token-act-pair · gpt-3.5-turbo

the word "stupid" and its variations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AUT

-0.93

apers

-0.83

 largeDownload

-0.83

APH

-0.82

accompan

-0.76

riott

-0.75

aver

-0.75

orthy

-0.74

rigan

-0.73

arnaev

-0.71

POSITIVE LOGITS

nesses

1.04

ly

0.95

ness

0.89

itude

0.81

gery

0.80

glers

0.77

 stupid

0.77

founded

0.77

ged

0.71

 shit

0.71

Activations Density 0.035%

words related to negative judgment of intelligence or actions

the word "stupid" and its variations in various contexts

No Comments

No Known Activations