INDEX

Explanations

specific contextual keywords and terms related to coding, legal matters, or academic language

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

Holiday Mansion, Policy online

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

principalColumn

-0.46

WindowConstants

-0.44

 StringTokenizer

-0.41

NameInMap

-0.40

 ویکی‌آمباردا

-0.38

WireFormatLite

-0.38

Partici

-0.37

arangay

-0.37

MeasureSpec

-0.36

</thead>

-0.34

POSITIVE LOGITS

 صوتيه

0.60

AddTagHelper

0.58

rrggbb

0.55

MigrationBuilder

0.54

 IFTT

0.48

PerformLayout

0.47

 ControllerBase

0.46

 Moyen

0.45

sauce

0.43

 autorytatywna

0.42

Activations Density 0.101%

specific contextual keywords and terms related to coding, legal matters, or academic language

Holiday Mansion, Policy online

No Comments

No Known Activations