INDEX

Explanations

conditional expressions or statements indicating checks

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

conditional negation

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

 sphere

-0.54

起来

-0.47

 canst

-0.47

 Sphere

-0.47

 Umgang

-0.46

sphere

-0.45

 approach

-0.45

起來

-0.45

 nationales

-0.44

affaires

-0.43

POSITIVE LOGITS

(!

1.07

(!

0.99

(!__

0.96

((!

0.76

(!_

0.75

(!$

0.71

(!$

0.71

(!_

0.66

posedge

0.66

assertFalse

0.65

Activations Density 0.009%

conditional expressions or statements indicating checks

conditional negation

No Comments

No Known Activations