INDEX

Explanations

ranges of numerical values specified with a hyphen

oai_token-act-pair · gpt-3.5-turbo

negative percentage ranges or numeric values associated with measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 wedd

-0.68

 Bates

-0.66

 Nasa

-0.65

 Dickinson

-0.63

 Cato

-0.63

Tus

-0.62

Cec

-0.62

 Defenders

-0.61

 Bryan

-0.61

 Cassidy

-0.61

POSITIVE LOGITS

sided

1.16

hour

1.15

year

1.15

percent

1.14

years

1.10

degree

1.09

minute

1.06

level

1.06

dimensional

1.05

digit

1.05

Activations Density 0.095%

ranges of numerical values specified with a hyphen

negative percentage ranges or numeric values associated with measurements

No Comments

No Known Activations