INDEX

Explanations

specific actions, processes, or adjustments related to modules, events, or situational changes

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

operation is performed

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

AndEndTag

-0.52

 Exacts

-0.47

reserv

-0.45

براير

-0.42

guiente

-0.42

guantes

-0.41

 Reviewing

-0.40

paisaje

-0.40

colgante

-0.40

 indígen

-0.39

POSITIVE LOGITS

BeginContext

0.45

 հղումներ

0.41

GraphicsUnit

0.40

estacks

0.40

 Lauder

0.39

плуата

0.39

odeon

0.37

 autorytatywna

0.37

 Chwiliwch

0.36

RefNanny

0.36

Activations Density 0.146%

specific actions, processes, or adjustments related to modules, events, or situational changes

operation is performed

No Comments

No Known Activations