INDEX

Explanations

references to statistical terms like "standard deviation."

oai_token-act-pair · gpt-3.5-turbo

references to statistical terms related to standard deviation and standardization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

frey

-0.75

lez

-0.75

mers

-0.72

pots

-0.71

activate

-0.70

amia

-0.69

gin

-0.68

leans

-0.68

ritz

-0.67

les

-0.66

POSITIVE LOGITS

 deviation

1.57

 deviations

1.42

 bearer

1.39

ization

1.33

ised

1.29

isation

1.11

ized

1.03

izations

1.02

izing

1.02

izes

1.00

Activations Density 0.051%

references to statistical terms like "standard deviation."

references to statistical terms related to standard deviation and standardization

No Comments

No Known Activations