INDEX

Explanations

structures and coding terminology

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

`const` and `ENOT` keywords

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

enso

-0.48

Personendaten

-0.44

 consultato

-0.44

tagHelperRunner

-0.44

 BoxDecoration

-0.43

!*\

-0.43

mensagem

-0.43

putExtra

-0.43

OCCURRED

-0.43

Sna

-0.41

POSITIVE LOGITS

 ainfi

0.53

 Juifs

0.47

 himſelf

0.47

0.45

 canst

0.45

 onely

0.44

 cuarzo

0.43

Activations Density 0.061%

structures and coding terminology

`const` and `ENOT` keywords

No Comments

No Known Activations

structures and coding terminology

`const` and `ENOT` keywords

No Comments

No Known Activations