INDEX

Explanations

phrases indicating a gap or deficiency in knowledge or resources

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

filling gaps and voids

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

-0.30

run

-0.29

toCharArray

-0.28

แพ

-0.27

-0.26

ത

-0.26

 manner

-0.26

ji

-0.26

Con

-0.25

 style

-0.25

POSITIVE LOGITS

 Gaps

1.07

 gaps

1.06

gaps

0.93

Gap

0.92

gap

0.87

gap

0.84

Gap

0.84

IntoConstraints

0.78

filling

0.77

Filling

0.76

Activations Density 0.017%

phrases indicating a gap or deficiency in knowledge or resources

filling gaps and voids

No Comments

No Known Activations