INDEX

Explanations

quotes or phrases encapsulating strong directives or beliefs

oai_token-act-pair · gpt-3.5-turbo

quotation marks in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Ļ½

-0.74

²¾

-0.71

psc

-0.66

İĭ

-0.65

¾

-0.64

MpServer

-0.64

»Ĵ

-0.63

itial

-0.62

acements

-0.62

emis

-0.60

POSITIVE LOGITS

/"

1.00

([

0.79

 advised

0.75

["

0.74

 âĢķ

0.69

(*

0.68

 âĨĴ

0.66

 blah

0.66

 justifies

0.65

 concludes

0.65

Activations Density 0.085%

quotes or phrases encapsulating strong directives or beliefs

quotation marks in the text

No Comments

No Known Activations