INDEX

Explanations

phrases indicating certainty or evidence

oai_token-act-pair · gpt-3.5-turbo

phrases that refer to signs or evidence of trends or outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

zanne

-0.77

ille

-0.70

ps

-0.68

zzo

-0.64

 Klux

-0.64

tz

-0.64

olly

-0.63

pered

-0.63

zes

-0.62

java

-0.62

POSITIVE LOGITS

 indication

1.11

 indications

1.07

 signs

0.93

 Signs

0.91

 indicators

0.85

 hint

0.84

terday

0.82

 indicator

0.81

 hints

0.80

orial

0.79

Activations Density 0.013%

phrases indicating certainty or evidence

phrases that refer to signs or evidence of trends or outcomes

No Comments

No Known Activations