INDEX

Explanations

references to lists or formats containing structured data elements

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

Admirals

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

/*---

-0.57

 kasarigan

-0.47

googleapis

-0.47

formik

-0.44

velocityY

-0.43

 nahilalakip

-0.42

ことができました

-0.41

 Wolken

-0.40

 Dodson

-0.40

蕩

-0.40

POSITIVE LOGITS

 ре

1.76

Ре

1.16

 Ре

1.15

ре

0.80

 РЕ

0.73

Re

0.71

Re

0.66

URLException

0.63

 Reggie

0.61

ERe

0.60

Activations Density 0.003%

references to lists or formats containing structured data elements

Admirals

No Comments

No Known Activations