INDEX

Explanations

phrases indicating uncertainty or lack of knowledge

oai_token-act-pair · gpt-3.5-turbo

phrases starting with "All" to emphasize totality or completeness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

•All the Features!

No Comments

Negative Logits

dl

-0.68

EStream

-0.64

rir

-0.63

argon

-0.63

IDS

-0.62

kamp

-0.59

ritic

-0.58

atro

-0.57

hole

-0.56

ciation

-0.56

POSITIVE LOGITS

owing

1.36

iances

1.32

iance

1.24

ocate

1.15

ocation

1.09

ergic

1.09

ocating

1.07

owed

1.04

yson

1.00

 bets

0.98

Activations Density 0.077%

phrases indicating uncertainty or lack of knowledge

phrases starting with "All" to emphasize totality or completeness

No Comments

No Known Activations