INDEX

Explanations

mentions of the word "oct" followed by a single digit number

oai_token-act-pair · gpt-3.5-turbo

references to the term "oct."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Interstitial

-0.84

 Limbaugh

-0.81

adder

-0.76

ACTED

-0.76

ndra

-0.75

HCR

-0.75

ãģĦ

-0.74

 Shapiro

-0.67

WARE

-0.67

ARP

-0.67

POSITIVE LOGITS

oct

1.37

oct

1.34

opus

1.14

ober

0.99

 Seym

0.96

uple

0.93

agonal

0.92

acles

0.83

agon

0.81

asus

0.80

Activations Density 0.008%

mentions of the word "oct" followed by a single digit number

references to the term "oct."

No Comments

No Known Activations