INDEX

Explanations

terms related to physical or mental effort

oai_token-act-pair · gpt-3.5-turbo

instances of the word "work" in various contexts related to effort and collaboration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ylon

-0.83

 Bubble

-0.71

 Ukrain

-0.71

anamo

-0.70

DragonMagazine

-0.69

netflix

-0.65

 Flavoring

-0.65

emonic

-0.65

antha

-0.64

 constitu

-0.62

POSITIVE LOGITS

 ethic

1.19

flows

0.99

hops

0.95

bench

0.92

horse

0.89

 tirelessly

0.87

manship

0.86

heet

0.83

 harder

0.83

 diligently

0.83

Activations Density 0.073%

terms related to physical or mental effort

instances of the word "work" in various contexts related to effort and collaboration

No Comments

No Known Activations