INDEX

Explanations

mentions of military organizations or activities

oai_token-act-pair · gpt-3.5-turbo

references to various military forces or armies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Consumer

-0.81

 Booker

-0.80

Bloom

-0.80

é¾įå¥ĳå£«

-0.76

Hop

-0.76

 Mormonism

-0.75

California

-0.73

 Bloom

-0.73

Lot

-0.70

 Neph

-0.69

POSITIVE LOGITS

 stationed

1.12

 colonel

0.99

YPG

0.99

 commanders

0.98

 commander

0.97

merga

0.91

 dispatched

0.91

 patrols

0.90

 deployed

0.89

 garrison

0.88

Activations Density 0.072%

mentions of military organizations or activities

references to various military forces or armies

No Comments

No Known Activations