INDEX

Explanations

specific references to documents and data records

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

United States, past due

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

tagHelperRunner

-0.82

majánló

-0.78

MigrationBuilder

-0.73

 uLocal

-0.73

 الرياضيه

-0.71

<unused68>

-0.71

<unused28>

-0.71

<unused41>

-0.71

<unused3>

-0.70

<unused8>

-0.70

POSITIVE LOGITS

 altså

0.29

 biały

0.27

 Así

0.27

 alltså

0.27

 अलावा

0.26

Así

0.26

 étoit

0.25

得

0.25

<h3>

0.25

至于

0.25

Activations Density 0.003%

specific references to documents and data records

United States, past due

No Comments

No Known Activations