INDEX

Explanations

legal and political terms and occurrences

oai_token-act-pair · gpt-3.5-turbo

references to articles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ascus

-0.87

adows

-0.79

nesota

-0.79

ITH

-0.75

ergic

-0.74

aukee

-0.73

cffff

-0.73

etsk

-0.71

ixtape

-0.70

escal

-0.69

POSITIVE LOGITS

ICLE

0.81

 Articles

0.80

meal

0.79

 Continued

0.78

 Article

0.75

Article

0.71

Header

0.70

hook

0.68

 Clicker

0.68

 Consent

0.67

Activations Density 0.015%

legal and political terms and occurrences

references to articles

No Comments

No Known Activations