INDEX

Explanations

elements related to decision-making or governance

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

after any gain

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

 CONCLUS

-0.43

anskje

-0.42

dighet

-0.41

 poveznice

-0.41

 Anschluss

-0.39

rilis

-0.38

 Fenster

-0.38

 Knoblauch

-0.38

 Federico

-0.38

 pascua

-0.38

POSITIVE LOGITS

RTEE

0.47

 OFDb

0.47

Autoritní

0.44

 Савезне

0.43

principalTable

0.42

WebServlet

0.41

 للاسماء

0.41

adaptiveStyles

0.40

GOTREF

0.40

!*\

0.39

Activations Density 0.214%

elements related to decision-making or governance

after any gain

No Comments

No Known Activations