INDEX

Explanations

phrases or words indicating uncertainty or speculation

oai_token-act-pair · gpt-3.5-turbo

phrases that suggest alternatives or comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ires

-0.72

pire

-0.65

ETS

-0.63

urs

-0.62

edia

-0.62

efer

-0.62

hesive

-0.61

mitter

-0.60

ailable

-0.60

moil

-0.59

POSITIVE LOGITS

acle

1.29

acles

1.27

chard

1.21

nam

1.21

acular

1.18

ifice

1.15

chid

1.13

 whatever

1.12

Else

1.11

whatever

1.07

Activations Density 0.185%

phrases or words indicating uncertainty or speculation

phrases that suggest alternatives or comparisons

No Comments

No Known Activations