INDEX

Explanations

words related to conflict and strained relationships in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to conflict or strain in relationships or situations

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

cise

-0.80

prints

-0.78

Interstitial

-0.72

icle

-0.72

sites

-0.72

glas

-0.68

 nominated

-0.68

icles

-0.68

cause

-0.67

metic

-0.66

POSITIVE LOGITS

 tension

1.00

 tensions

0.95

 reliever

0.85

 cooker

0.80

 flared

0.78

yip

0.74

PLA

0.70

 Vaugh

0.69

 guiActiveUn

0.69

atile

0.69

Activations Density 0.022%

words related to conflict and strained relationships in various contexts

references to conflict or strain in relationships or situations

No Comments

No Known Activations