INDEX

Explanations

verbs or phrases related to secretive or deceitful behavior

oai_token-act-pair · gpt-3.5-turbo

terms related to collusion and exclusion

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

@#&

-0.83

 rainy

-0.72

pora

-0.68

Ü

-0.67

 reinvest

-0.63

 Trials

-0.60

 sunset

-0.59

awk

-0.58

Hop

-0.58

Rai

-0.58

POSITIVE LOGITS

uding

1.06

udes

1.03

uded

1.01

ebus

0.94

geoning

0.89

ollah

0.86

ude

0.85

enegger

0.77

ociate

0.77

ctuary

0.75

Activations Density 0.007%

verbs or phrases related to secretive or deceitful behavior

terms related to collusion and exclusion

No Comments

No Known Activations