INDEX

Explanations

references to study participants or research participants

oai_token-act-pair · gpt-3.5-turbo

mentions of study participants in the context of research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 vengeance

-0.75

RAY

-0.64

 thicker

-0.60

 Shack

-0.57

 exceptions

-0.56

bey

-0.55

away

-0.55

oÄŁ

-0.55

unts

-0.55

 pastoral

-0.54

POSITIVE LOGITS

 participant

1.00

 Participants

1.00

 participants

0.92

cript

0.91

 Participant

0.90

particip

0.88

 Participation

0.77

izers

0.76

Particip

0.76

insula

0.75

Activations Density 0.015%

references to study participants or research participants

mentions of study participants in the context of research

No Comments

No Known Activations