INDEX

Explanations

programming languages and related syntax elements like hashtags

oai_token-act-pair · gpt-3.5-turbo

instances of the character "#" used in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Gors

-0.74

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.73

vation

-0.72

ator

-0.67

 EntityItem

-0.66

tsky

-0.65

Vie

-0.65

sew

-0.64

 Karin

-0.63

 seeker

-0.63

POSITIVE LOGITS

define

1.10

!/

1.08

endif

1.03

wcsstore

1.02

DonaldTrump

0.92

abuse

0.89

DIV

0.86

saf

0.83

users

0.82

cdn

0.81

Activations Density 0.013%

programming languages and related syntax elements like hashtags

instances of the character "#" used in various contexts

No Comments

No Known Activations