INDEX

Explanations

words related to time or temporal proximity

oai_token-act-pair · gpt-3.5-turbo

references to impending future events or conditions

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

ensional

-0.53

 Canaver

-0.50

ctions

-0.48

 Matter

-0.46

sect

-0.45

 Bosh

-0.45

 Invention

-0.44

 Pens

-0.44

casters

-0.44

[+

-0.43

POSITIVE LOGITS

an

0.67

aned

0.62

\\

0.59

ana

0.58

ans

0.58

ane

0.55

inges

0.54

lihood

0.54

ja

0.53

aning

0.52

Activations Density 0.599%

words related to time or temporal proximity

references to impending future events or conditions

No Comments

No Known Activations