INDEX

Explanations

references to or mentions of the country or region 'India'

oai_token-act-pair · gpt-3.5-turbo

references to India or Indian-related topics

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sonian

-0.87

thodox

-0.81

EStream

-0.80

=-=-=-=-=-=-=-=-

-0.73

isine

-0.72

acters

-0.71

vre

-0.69

olitan

-0.69

phy

-0.69

bley

-0.68

POSITIVE LOGITS

ividually

1.15

irect

1.13

ebted

1.07

epend

0.99

icative

0.98

enture

0.96

Ind

0.92

Ind

0.91

icator

0.90

icates

0.86

Activations Density 0.006%

references to or mentions of the country or region 'India'

references to India or Indian-related topics

No Comments

No Known Activations