INDEX

Explanations

programming structure and syntax elements

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

public class members

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

WebElementEntity

-0.68

AndEndTag

-0.57

:✨

-0.54

MessageOf

-0.49

RegressionTest

-0.49

ագրություններ

-0.48

 ujednoznacz

-0.48

يكب

-0.47

PreferredItem

-0.46

SBATCH

-0.46

POSITIVE LOGITS

mar

0.40

pter

0.39

oodle

0.38

pleen

0.38

Pie

0.35

user

0.35

 Proven

0.35

脚注の使い方

0.35

mom

0.35

tag

0.35

Activations Density 0.011%

programming structure and syntax elements

public class members

No Comments

No Known Activations