INDEX

Explanations

references to family members, particularly sons

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "son."

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

veyard

-0.70

 freight

-0.66

Union

-0.65

iculty

-0.64

PORT

-0.63

icial

-0.59

ords

-0.59

 manifold

-0.59

 accommodations

-0.59

TING

-0.58

POSITIVE LOGITS

 Gohan

0.97

hood

0.91

nets

0.88

hesis

0.87

ogram

0.87

pins

0.83

hetically

0.83

mares

0.79

son

0.79

 Barron

0.78

Activations Density 0.025%

references to family members, particularly sons

mentions of the word "son."

No Comments

No Known Activations