INDEX

Explanations

phrases indicating a comparison or contrast

oai_token-act-pair · gpt-3.5-turbo

phrases indicating reliance or attribution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ramids

-0.74

ongo

-0.65

 chemistry

-0.64

Mem

-0.64

 anonymity

-0.63

uku

-0.61

iker

-0.61

mill

-0.61

artifacts

-0.60

MQ

-0.60

POSITIVE LOGITS

 tallest

0.68

 mattered

0.68

 longest

0.66

 Carney

0.65

abulary

0.64

 escaped

0.64

kson

0.62

soever

0.62

dn

0.62

 Nare

0.61

Activations Density 0.143%

phrases indicating a comparison or contrast

phrases indicating reliance or attribution

No Comments

No Known Activations