INDEX

Explanations

numerical values and technical terms related to medical data and experiments

oai_token-act-pair · gpt-3.5-turbo

statistical data related to medical studies or experiments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ODY

-0.85

EStream

-0.78

DragonMagazine

-0.70

å§«

-0.68

Loading

-0.66

dale

-0.66

itably

-0.65

lot

-0.65

mite

-0.65

zeb

-0.65

POSITIVE LOGITS

Â±

1.13

/-

0.96

 Distribut

0.69

Gur

0.63

Vaj

0.62

ribut

0.62

0.60

 Indones

0.60

ng

0.60

riks

0.59

Activations Density 0.002%

numerical values and technical terms related to medical data and experiments

statistical data related to medical studies or experiments

No Comments

No Known Activations