INDEX

Explanations

references to the 9/11 attacks

oai_token-act-pair · gpt-3.5-turbo

references to the 9/11 attacks

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

Tile

-0.70

eva

-0.67

hunt

-0.64

maid

-0.63

 pengu

-0.63

Jem

-0.63

laus

-0.62

Nun

-0.62

chance

-0.62

 unman

-0.62

POSITIVE LOGITS

 Syndrome

0.84

 anniversary

0.83

 Truth

0.76

abad

0.76

 bombers

0.74

amara

0.74

truth

0.72

iets

0.72

 bombings

0.72

Truth

0.71

Activations Density 0.024%

references to the 9/11 attacks

references to the 9/11 attacks

No Comments

No Known Activations