INDEX

Explanations

mentions of people's names or words related to their positions or titles

oai_token-act-pair · gpt-3.5-turbo

references to the concept of leadership or authority

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

 decomp

-0.90

 JPEG

-0.80

 liter

-0.73

 silhou

-0.64

 compression

-0.63

 scrut

-0.63

jog

-0.63

 whistle

-0.63

 vulner

-0.63

 Palestin

-0.62

POSITIVE LOGITS

1.39

ship

0.97

swer

0.94

tal

0.92

ski

0.90

sen

0.90

alty

0.89

sa

0.88

ships

0.87

tre

0.85

Activations Density 0.257%

mentions of people's names or words related to their positions or titles

references to the concept of leadership or authority

No Comments

No Known Activations