INDEX

Explanations

phrases related to computer programming

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and their contextual use in listing or elaborating points

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

ctuary

-0.65

velop

-0.65

eport

-0.64

onom

-0.63

Thirty

-0.61

ublished

-0.59

omal

-0.59

inguishable

-0.59

itialized

-0.58

nex

-0.58

POSITIVE LOGITS

etc

1.20

etc

1.04

 blah

1.02

but

1.00

BUT

0.96

 respectively

0.95

but

0.94

 whereas

0.92

 Anyway

0.86

 haha

0.85

Activations Density 0.526%

phrases related to computer programming

punctuation marks and their contextual use in listing or elaborating points

No Comments

No Known Activations