INDEX

Explanations

numerical values and references to time periods

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

years and dates

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

amssymb

-0.47

 stället

-0.45

 للمعارف

-0.45

Skocz

-0.45

الإنجليزية

-0.44

PerformLayout

-0.44

msgTypes

-0.43

 ویکی‌پدی

-0.42

 Demonstration

-0.42

 gonzález

-0.42

POSITIVE LOGITS

 actionMode

0.44

ngdoc

0.42

 betweenstory

0.40

ulfon

0.38

 épaules

0.35

 &___

0.35

 pandemi

0.34

 JAXB

0.34

+#+

0.34

VersionUID

0.33

Activations Density 0.019%

numerical values and references to time periods

years and dates

No Comments

No Known Activations