INDEX

Explanations

words related to conflict or stress

oai_token-act-pair · gpt-3.5-turbo

instances or mentions of tension in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

missions

-0.78

cise

-0.78

glas

-0.78

sites

-0.77

apple

-0.74

prints

-0.72

icle

-0.70

Interstitial

-0.70

lish

-0.69

look

-0.69

POSITIVE LOGITS

 tension

1.26

 tensions

1.09

 reliever

0.94

 guiActiveUn

0.89

yip

0.79

 escalation

0.78

 cooker

0.78

 friction

0.76

 Vaugh

0.75

 pneum

0.74

Activations Density 0.013%

words related to conflict or stress

instances or mentions of tension in various contexts

No Comments

No Known Activations