INDEX

Explanations

instances of programming-related terms and code snippets

oai_token-act-pair · gpt-3.5-turbo

references to social media and technology-related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

itably

-0.72

othal

-0.68

ividual

-0.67

irlf

-0.65

;;;;;;;;;;;;

-0.65

ogun

-0.64

ancial

-0.63

Ã©e

-0.59

uilt

-0.59

ogo

-0.59

POSITIVE LOGITS

CLASSIFIED

1.03

 antioxid

0.90

 behavi

0.87

 Seym

0.85

Balt

0.85

 suscept

0.84

NetMessage

0.83

WAYS

0.83

Interstitial

0.83

 welf

0.83

Activations Density 2.910%

instances of programming-related terms and code snippets

references to social media and technology-related concepts

No Comments

No Known Activations