INDEX

Explanations

the word "speaker"

oai_token-act-pair · gpt-3.5-turbo

references to speakers at events or conferences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

laus

-0.70

rises

-0.68

YP

-0.65

 Devils

-0.63

jac

-0.63

 Cron

-0.63

DIV

-0.61

 Reporting

-0.60

olitan

-0.58

utsu

-0.58

POSITIVE LOGITS

 speaker

1.42

 speakers

1.19

 Speaker

0.86

phone

0.85

loo

0.82

speech

0.78

voice

0.78

 Gingrich

0.77

 louder

0.77

spe

0.76

Activations Density 0.011%

the word "speaker"

references to speakers at events or conferences

No Comments

No Known Activations