INDEX

Explanations

phrases relating to global impact or significance

oai_token-act-pair · gpt-3.5-turbo

references to "the world."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

-0.85

imon

-0.85

ippi

-0.79

ients

-0.78

urations

-0.77

ibility

-0.75

å£«

-0.73

merce

-0.72

WARE

-0.72

andre

-0.72

POSITIVE LOGITS

wide

0.90

 premiere

0.84

 renowned

0.76

 wide

0.76

 Wide

0.74

 economy

0.73

 renown

0.70

 champion

0.69

eers

0.68

 premier

0.68

Activations Density 0.061%

phrases relating to global impact or significance

references to "the world."

No Comments

No Known Activations