INDEX

Explanations

references to unsigned and signed code

oai_token-act-pair · gpt-3.5-turbo

terms related to software application security and code signing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

borough

-0.93

rir

-0.90

Ã©n

-0.86

atre

-0.85

ologies

-0.84

ologic

-0.84

combe

-0.83

ologically

-0.83

aunders

-0.81

omon

-0.79

POSITIVE LOGITS

 unsigned

1.10

 signed

0.79

signed

0.78

 integers

0.73

 indemn

0.71

 uint

0.70

 hippocamp

0.70

 signing

0.69

unsigned

0.68

quo

0.68

Activations Density 0.005%

references to unsigned and signed code

terms related to software application security and code signing

No Comments

No Known Activations