INDEX

Explanations

dates written in numerical form

oai_token-act-pair · gpt-3.5-turbo

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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 contrace

-0.71

ilic

-0.64

iage

-0.64

rolet

-0.63

vati

-0.62

perty

-0.61

ainer

-0.60

 Kessler

-0.60

 spread

-0.60

isphere

-0.59

POSITIVE LOGITS

 SHARES

1.02

 Minutes

0.89

nd

0.88

 mins

0.81

ÃĹ

0.80

rd

0.79

bp

0.79

hrs

0.78

 Months

0.78

dB

0.77

Activations Density 0.142%

dates written in numerical form

numerical data or statistics

No Comments

No Known Activations