INDEX

Explanations

scientific studies, research findings, and academic papers

oai_token-act-pair · gpt-3.5-turbo

academic research and analyses of literary works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 mute

-0.56

 retaliate

-0.53

boss

-0.52

zz

-0.51

engers

-0.51

ansas

-0.51

 wink

-0.50

Shop

-0.50

stop

-0.49

bucks

-0.49

POSITIVE LOGITS

 methodological

1.19

 empirical

1.11

 analyses

1.07

 observational

1.03

doi

1.03

 hypotheses

1.02

Abstract

1.01

 studies

1.01

Study

1.00

 study

0.99

Activations Density 0.668%

scientific studies, research findings, and academic papers

academic research and analyses of literary works

No Comments

No Known Activations