INDEX

Explanations

expressions of near-completeness or situations that are close to being true

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 lugs

-0.72

RefreshLayout

-0.62

ṇ

-0.61

 culoare

-0.59

Slf

-0.59

 Silverman

-0.59

Spinner

-0.59

 picioare

-0.58

 Guen

-0.58

POSITIVE LOGITS

 nearly

1.31

 almost

1.20

Casi

1.19

 Almost

1.15

Almost

1.15

 Nearly

1.14

nearly

1.13

 Casi

1.08

Nearly

1.07

almost

1.02

Activations Density 0.074%

expressions of near-completeness or situations that are close to being true

No Comments

No Known Activations