INDEX

Explanations

references to explosive devices

oai_token-act-pair · gpt-3.5-turbo

references to bombs and explosive devices

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

laus

-0.93

SEA

-0.79

éĸ

-0.75

é»Ĵ

-0.69

Past

-0.69

 Vale

-0.69

 Legends

-0.68

ç¥ŀ

-0.66

DAY

-0.66

Lic

-0.65

POSITIVE LOGITS

arding

1.28

arded

1.17

 detonated

1.06

 deton

0.97

astic

0.93

 bomb

0.93

 explodes

0.91

astically

0.91

 bombs

0.89

 bombings

0.84

Activations Density 0.015%

references to explosive devices

references to bombs and explosive devices

No Comments

No Known Activations