INDEX

Explanations

time-related words conveying a sense of comparison or change over time

oai_token-act-pair · gpt-3.5-turbo

temporal references and comparisons related to time intervals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

inite

-0.68

itely

-0.61

quartered

-0.60

jac

-0.59

Fit

-0.56

cientious

-0.56

olate

-0.54

ancial

-0.54

ospace

-0.54

ocaly

-0.52

POSITIVE LOGITS

ago

1.83

 earlier

1.41

Ago

1.37

 prior

1.15

 previously

1.13

 previous

1.03

ago

1.01

 before

0.92

 beforehand

0.89

 predecessors

0.88

Activations Density 0.065%

time-related words conveying a sense of comparison or change over time

temporal references and comparisons related to time intervals

No Comments

No Known Activations