INDEX

Explanations

references to data and signals in a technical or scientific context

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

subsequent steps

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

endregion

-0.40

AutoScaleMode

-0.39

TÜ

-0.39

ZZLE

-0.39

thwaite

-0.38

ConverterFactory

-0.38

 GetEnumerator

-0.38

TabStop

-0.37

組

-0.37

/*---

-0.36

POSITIVE LOGITS

tagHelperRunner

0.50

vė

0.46

 Daarna

0.45

 Anschließend

0.45

 afterwards

0.45

 násled

0.45

 anschließend

0.43

Jeografia

0.43

istoitu

0.43

 Italijani

0.42

Activations Density 0.171%

references to data and signals in a technical or scientific context

subsequent steps

No Comments

No Known Activations