INDEX

Explanations

instances of the word "master."

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

master in technical contexts

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

Hentet

-0.65

 للمعارف

-0.58

afficheront

-0.54

 increí

-0.51

 proyección

-0.47

 incrí

-0.47

achting

-0.46

 indígen

-0.46

 Italijani

-0.46

lgari

-0.45

POSITIVE LOGITS

master

2.05

 master

1.61

Master

1.53

MASTER

1.48

 Master

1.48

 MASTER

1.35

masters

1.28

 masters

1.16

マスター

1.10

 Masters

1.05

Activations Density 0.002%

instances of the word "master."

master in technical contexts

No Comments

No Known Activations