INDEX

Explanations

references to captions in online media articles

oai_token-act-pair · gpt-3.5-turbo

phrases related to counts and statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

tracks

-0.96

Virgin

-0.93

Ak

-0.88

Liber

-0.83

skirts

-0.83

Cu

-0.83

Cass

-0.83

BAT

-0.81

Rush

-0.81

bah

-0.81

POSITIVE LOGITS

 bloc

0.71

MLS

0.70

 poll

0.69

toe

0.68

 ÂŃ

0.66

 quarter

0.66

 bracket

0.64

 decimal

0.63

tip

0.63

 batch

0.62

Activations Density 0.114%

references to captions in online media articles

phrases related to counts and statistics

No Comments

No Known Activations