INDEX

Explanations

descriptive elements related to settings and environments

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

genodigd

-0.49

 kasarigan

-0.45

 resultaat

-0.43

 gebru

-0.41

 personalization

-0.41

 radiological

-0.41

 StringTokenizer

-0.40

 belangrijkste

-0.40

gitto

-0.40

obatan

-0.39

POSITIVE LOGITS

 autorytatywna

0.53

 surla

0.52

LookAnd

0.44

Autoritní

0.44

IsContent

0.41

 living

0.39

 TextAppearance

0.39

AutoScaleMode

0.37

 AssemblyVersion

0.37

 AttributeSet

0.36

Activations Density 0.043%

descriptive elements related to settings and environments

No Comments

No Known Activations