INDEX

Explanations

words related to strong emotional reactions or intensity

oai_token-act-pair · gpt-3.5-turbo

words related to significant legal terms and conditions

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

spr

-0.94

sweet

-0.78

jack

-0.74

lance

-0.70

lesi

-0.70

FFFF

-0.68

nah

-0.66

els

-0.66

Spr

-0.66

eming

-0.66

POSITIVE LOGITS

 metic

0.94

atically

0.84

 guiActiveUn

0.84

ixel

0.80

othing

0.80

othe

0.78

 captcha

0.77

ographic

0.74

 ancest

0.74

orical

0.71

Activations Density 0.067%

words related to strong emotional reactions or intensity

words related to significant legal terms and conditions

No Comments

No Known Activations