INDEX

Explanations

instances related to digital operations, such as commands and processes

oai_token-act-pair · gpt-3.5-turbo

nouns and entities related to various categories and occurrences

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yss

-0.70

ç·

-0.70

cgi

-0.68

reditary

-0.66

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.64

cffffcc

-0.62

xious

-0.62

çī

-0.62

åŃĲ

-0.61

sonian

-0.61

POSITIVE LOGITS

 totaling

1.01

 apiece

1.00

hips

0.95

mith

0.94

 spanning

0.93

 imaginable

0.92

 including

0.89

 vying

0.88

 dotted

0.87

 ranging

0.86

Activations Density 0.432%

instances related to digital operations, such as commands and processes

nouns and entities related to various categories and occurrences

No Comments

No Known Activations