INDEX

Explanations

mentions of or support for political or military intervention

oai_token-act-pair · gpt-3.5-turbo

terms related to government or military intervention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Ĥ¬

-0.68

fab

-0.65

scribe

-0.61

 Bender

-0.60

union

-0.60

 Bones

-0.60

guyen

-0.60

 Burr

-0.60

house

-0.60

hend

-0.59

POSITIVE LOGITS

 milit

0.94

 intervene

0.90

ateral

0.85

 interven

0.78

 intervention

0.78

aries

0.77

ist

0.76

 intervened

0.74

anke

0.73

ional

0.72

Activations Density 0.047%

mentions of or support for political or military intervention

terms related to government or military intervention

No Comments

No Known Activations