INDEX

Explanations

JavaScript code snippets and related key terms

oai_token-act-pair · gpt-3.5-turbo

technical terms and programming constructs

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

 arteries

-0.75

 resilience

-0.68

 Lutheran

-0.66

 athe

-0.65

 citiz

-0.64

ouver

-0.62

 adherence

-0.62

 grit

-0.61

deen

-0.60

ĪĴ

-0.59

POSITIVE LOGITS

-$

1.00

Container

0.82

à¤¾

0.71

åħī

0.71

adr

0.71

#$#$

0.70

Ĥİ

0.70

JSON

0.67

DragonMagazine

0.65

WithNo

0.65

Activations Density 0.102%

JavaScript code snippets and related key terms

technical terms and programming constructs

No Comments

No Known Activations