INDEX

Explanations

news or article corrections or updates

oai_token-act-pair · gpt-3.5-turbo

references to preceding or updated versions of a story

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

emale

-0.85

ignt

-0.84

alus

-0.84

aez

-0.83

ojure

-0.81

amily

-0.81

ategory

-0.79

efficients

-0.79

ajor

-0.78

sterdam

-0.78

POSITIVE LOGITS

telling

1.02

 revolving

0.93

board

0.88

arc

0.87

boards

0.86

 reprinted

0.86

book

0.82

 REPORT

0.79

fare

0.79

 arcs

0.79

Activations Density 0.041%

news or article corrections or updates

references to preceding or updated versions of a story

No Comments

No Known Activations