INDEX

Explanations

words related to violent events or conflicts

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

phrases indicating casualties or injuries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

mination

-0.74

SpaceEngineers

-0.74

isSpecial

-0.73

ThumbnailImage

-0.71

dayName

-0.69

ãĤ´ãĥ³

-0.69

ħĭ

-0.68

ociation

-0.66

StreamerBot

-0.64

ă

-0.64

POSITIVE LOGITS

0.88

0.87

0.86

 nine

0.85

 eight

0.84

Activations Density 0.102%

words related to violent events or conflicts

phrases indicating casualties or injuries

No Comments

No Known Activations