INDEX

Explanations

numeric data or values

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 soñar

-0.47

 cómod

-0.44

ismet

-0.44

 surla

-0.43

MigrationBuilder

-0.42

ясь

-0.42

 cómodo

-0.41

 SYLLABLE

-0.41

 esfuer

-0.40

 kvinna

-0.40

POSITIVE LOGITS

ViewImports

0.65

帖最后由

0.54

 Autorizaciones

0.52

elemField

0.46

tonode

0.46

MessageState

0.46

adaptiveStyles

0.45

 autorytatywna

0.44

ContentAsync

0.44

ScopeManager

0.42

Activations Density 0.000%

numeric data or values

No Comments

No Known Activations