INDEX

Explanations

commands related to modifying content or attributes

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

review the

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

UserScript

-0.44

 Muni

-0.41

rítica

-0.41

 Pohl

-0.41

pictureBox

-0.41

GTCX

-0.40

 Sakai

-0.40

 showcase

-0.40

tvb

-0.40

 anchored

-0.39

POSITIVE LOGITS

Derbyniad

0.50

 gyhoeddwyd

0.49

 pleaſure

0.48

FunctionFlags

0.47

">//

0.46

 myſelf

0.45

 ſind

0.43

帖最后由

0.43

 ujednoznacz

0.41

ſelf

0.41

Activations Density 0.000%

commands related to modifying content or attributes

review the

No Comments

No Known Activations