INDEX

Explanations

phrases related to hard work or effort

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "grind" or hard work

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

idas

-0.70

uana

-0.70

tymology

-0.69

amera

-0.68

ortium

-0.65

erva

-0.63

abet

-0.63

 Loft

-0.62

vironment

-0.59

CEPT

-0.58

POSITIVE LOGITS

 grind

1.02

ingly

0.93

stone

0.92

 Grind

0.85

aign

0.84

tro

0.82

 grinding

0.81

downs

0.78

rice

0.77

sheet

0.76

Activations Density 0.024%

phrases related to hard work or effort

references to the concept of "grind" or hard work

No Comments

No Known Activations