INDEX

Explanations

references to a specific entity, "Dor"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Dor."

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

anwhile

-0.95

WAYS

-0.75

unct

-0.65

HS

-0.64

 Terrorism

-0.64

)=(

-0.63

BRE

-0.63

tenance

-0.62

eers

-0.62

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.61

POSITIVE LOGITS

ado

0.99

iane

0.98

ÃŃa

0.97

je

0.93

oshenko

0.89

cas

0.89

omial

0.87

rell

0.87

chester

0.87

acies

0.87

Activations Density 0.012%

references to a specific entity, "Dor"

occurrences of the name "Dor."

No Comments

No Known Activations