INDEX

Explanations

words related to expressing strong opinions or dissatisfaction

oai_token-act-pair · gpt-3.5-turbo

various forms of the word "rant" and its related derivatives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

é¾įå

-0.73

Nir

-0.62

Bei

-0.62

enza

-0.61

ovo

-0.61

nai

-0.59

 captcha

-0.59

 Bonds

-0.59

olphins

-0.58

WIN

-0.58

POSITIVE LOGITS

kefeller

0.93

igious

0.81

imentary

0.78

backer

0.77

Edited

0.74

bery

0.74

naire

0.73

ishly

0.71

gow

0.70

essional

0.70

Activations Density 0.046%

words related to expressing strong opinions or dissatisfaction

various forms of the word "rant" and its related derivatives

No Comments

No Known Activations