INDEX

Explanations

terms related to the Air Force

oai_token-act-pair · gpt-3.5-turbo

mentions of the "Air Force."

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

enance

-0.74

 estab

-0.68

 guiIcon

-0.63

é¾įå¥ĳå£«

-0.63

avering

-0.63

oldemort

-0.63

 ç¥ŀ

-0.62

umbn

-0.62

fmt

-0.62

ection

-0.62

POSITIVE LOGITS

plane

1.10

planes

0.99

waves

0.94

raid

0.89

 Force

0.81

blast

0.81

Train

0.80

Air

0.80

Raid

0.80

ship

0.79

Activations Density 0.015%

terms related to the Air Force

mentions of the "Air Force."

No Comments

No Known Activations