INDEX

Explanations

spoken information or quotes

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Speaking" and its variations, indicating discussions or statements made by someone

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eer

-0.58

vier

-0.57

CoC

-0.57

involved

-0.55

jl

-0.55

 creation

-0.54

YP

-0.53

zip

-0.53

eers

-0.52

Pg

-0.52

POSITIVE LOGITS

 anonymously

1.10

 exclusively

0.95

 onstage

0.92

 afterwards

0.88

 candid

0.84

 separately

0.84

 shortly

0.83

 yesterday

0.83

 nostalg

0.83

 alongside

0.83

Activations Density 0.037%

spoken information or quotes

instances of the word "Speaking" and its variations, indicating discussions or statements made by someone

No Comments

No Known Activations