INDEX

Explanations

adjectives expressing opinions or evaluations about things or people

oai_token-act-pair · gpt-3.5-turbo

words related to characterization and opinions

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

ammy

-0.66

hoff

-0.64

cum

-0.64

nir

-0.64

perty

-0.62

neau

-0.60

stration

-0.60

externalActionCode

-0.59

kr

-0.59

isoft

-0.59

POSITIVE LOGITS

phas

0.94

 ourselves

0.83

 oneself

0.79

encies

0.78

it

0.77

Ī

0.77

enance

0.76

pointers

0.74

 themselves

0.74

tones

0.73

Activations Density 0.199%

adjectives expressing opinions or evaluations about things or people

words related to characterization and opinions

No Comments

No Known Activations