INDEX

Explanations

mentions of the word "university."

oai_token-act-pair · gpt-3.5-turbo

references to universities or higher education institutions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

othy

-0.85

trap

-0.75

adow

-0.73

ppings

-0.73

BOOK

-0.69

nir

-0.68

owder

-0.67

Jere

-0.66

avior

-0.64

riot

-0.64

POSITIVE LOGITS

 tuition

1.04

 campuses

1.03

uates

1.02

 lecturer

0.87

 campus

0.85

 educated

0.83

educated

0.83

 graduate

0.83

 fraternity

0.81

 graduates

0.80

Activations Density 0.026%

mentions of the word "university."

references to universities or higher education institutions

No Comments

No Known Activations