INDEX

Explanations

mentions of specific names related to politics, possibly in conflict regions

oai_token-act-pair · gpt-3.5-turbo

names of geographic regions, organizations, or prominent individuals associated with those areas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ENCY

-0.74

 Doodle

-0.67

 monkey

-0.66

ciplinary

-0.63

 Korean

-0.63

 Scorpion

-0.63

amus

-0.62

é¾įå¥ĳå£«

-0.62

 pies

-0.61

 Korea

-0.61

POSITIVE LOGITS

rina

0.85

dar

0.82

azard

0.81

rd

0.80

bian

0.79

raviolet

0.79

edly

0.78

rons

0.77

ril

0.75

iated

0.75

Activations Density 0.041%

mentions of specific names related to politics, possibly in conflict regions

names of geographic regions, organizations, or prominent individuals associated with those areas

No Comments

No Known Activations