INDEX

Explanations

call-to-action phrases related to subscribing or unsubscribing from a service

oai_token-act-pair · gpt-3.5-turbo

terms related to subscription and unsubscribing processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Pist

-0.67

ried

-0.66

berman

-0.66

orney

-0.63

dan

-0.62

doms

-0.60

BALL

-0.59

esan

-0.59

ritz

-0.58

pex

-0.58

POSITIVE LOGITS

Interstitial

0.81

taboola

0.81

Subscribe

0.71

 isEnabled

0.70

henko

0.69

ãĤ®

0.68

 Pastebin

0.65

 anymore

0.65

ickets

0.63

 securely

0.63

Activations Density 0.007%

call-to-action phrases related to subscribing or unsubscribing from a service

terms related to subscription and unsubscribing processes

No Comments

No Known Activations