INDEX

Explanations

phrases related to opting out or not participating

oai_token-act-pair · gpt-3.5-turbo

phrases related to opting out of services or agreements

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

GI

-0.63

ournals

-0.62

sworth

-0.62

Found

-0.60

 Liver

-0.59

 Gazette

-0.58

 fundament

-0.58

staff

-0.58

Sol

-0.57

gow

-0.56

POSITIVE LOGITS

aye

0.73

wagen

0.73

 bandwagon

0.70

issan

0.66

ersen

0.65

fitted

0.65

 parole

0.65

skirts

0.64

haps

0.63

 subscribing

0.62

Activations Density 0.024%

phrases related to opting out or not participating

phrases related to opting out of services or agreements

No Comments

No Known Activations