INDEX

Explanations

mentions of specific geopolitical entities or jurisdictions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating states or conditions, particularly in political or social contexts

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

ï¸ı

-0.73

bley

-0.71

iator

-0.65

 Cosponsors

-0.65

 Telescope

-0.63

comed

-0.63

ÃŃs

-0.62

Zup

-0.61

<?

-0.61

poon

-0.60

POSITIVE LOGITS

 Decay

0.77

foundland

0.77

 affairs

0.75

CLASSIFIED

0.73

IFIED

0.71

inence

0.69

ortunately

0.65

rency

0.65

vironment

0.65

gain

0.65

Activations Density 0.070%

mentions of specific geopolitical entities or jurisdictions

phrases indicating states or conditions, particularly in political or social contexts

No Comments

No Known Activations