INDEX

Explanations

phrases indicating investigation or consideration of a topic

oai_token-act-pair · gpt-3.5-turbo

the phrase "look into."

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

ienne

-0.65

hop

-0.63

 farewell

-0.62

bender

-0.62

warning

-0.60

owl

-0.60

END

-0.59

present

-0.58

 acquitted

-0.58

NES

-0.57

POSITIVE LOGITS

âĸĳâĸĳ

0.73

zag

0.69

 feasibility

0.68

 whats

0.66

 Coinbase

0.65

effic

0.63

 whether

0.62

izons

0.62

agall

0.62

 Divinity

0.62

Activations Density 0.026%

phrases indicating investigation or consideration of a topic

the phrase "look into."

No Comments

No Known Activations