INDEX

Explanations

proper nouns or names and potentially associated numbers

oai_token-act-pair · gpt-3.5-turbo

references to specific numerical data or statistics

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

xual

-0.80

undai

-0.74

ean

-0.73

phrine

-0.71

ĸļ

-0.70

mble

-0.67

pressure

-0.67

tumblr

-0.65

rawdownloadcloneembedreportprint

-0.65

irie

-0.64

POSITIVE LOGITS

asted

0.97

0.83

OTAL

0.81

asts

0.80

ature

0.79

_.

0.77

 Minutes

0.76

TI

0.76

ovember

0.75

dfx

0.75

Activations Density 0.028%

proper nouns or names and potentially associated numbers

references to specific numerical data or statistics

No Comments

No Known Activations