INDEX

Explanations

phrases related to comparing or contrasting different things

oai_token-act-pair · gpt-3.5-turbo

phrases indicating general consensus or common understanding

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

wic

-0.69

fug

-0.64

swers

-0.62

channelAvailability

-0.62

endas

-0.62

 offline

-0.61

etsk

-0.60

rex

-0.60

 Invaders

-0.60

itis

-0.59

POSITIVE LOGITS

 standards

1.31

 reckoning

1.18

 means

1.08

 accounts

1.02

 criteria

0.99

 Means

0.98

 definition

0.95

 measures

0.95

 criterion

0.92

stand

0.89

Activations Density 0.080%

phrases related to comparing or contrasting different things

phrases indicating general consensus or common understanding

No Comments

No Known Activations