INDEX

Explanations

words related to enhancement or improvement

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a positive contribution or improvement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

avior

-0.82

 Advertisement

-0.74

AAAAAAAA

-0.64

aneous

-0.62

cz

-0.61

irus

-0.60

DH

-0.60

unn

-0.60

CDC

-0.60

 apologized

-0.59

POSITIVE LOGITS

 humankind

0.74

otos

0.73

srf

0.72

 accompany

0.67

axy

0.67

tones

0.66

ggles

0.66

 compensate

0.66

 mankind

0.66

ound

0.65

Activations Density 0.133%

words related to enhancement or improvement

phrases indicating a positive contribution or improvement

No Comments

No Known Activations