INDEX

Explanations

words related to actions or processes

oai_token-act-pair · gpt-3.5-turbo

words related to various forms of "transition" or "conditioning."

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

ĻĤ

-0.70

psc

-0.65

solete

-0.65

WS

-0.62

aptic

-0.62

Recomm

-0.60

ä¹ĭ

-0.59

WN

-0.59

Neg

-0.59

YR

-0.58

POSITIVE LOGITS

ition

1.48

itious

1.16

itions

1.05

naire

0.91

icut

0.88

ality

0.77

eers

0.77

eer

0.76

eering

0.76

ITION

0.74

Activations Density 0.012%

words related to actions or processes

words related to various forms of "transition" or "conditioning."

No Comments

No Known Activations