INDEX

Explanations

evidence and indicators of associations in various contexts

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

statements of fact or relationship

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

 fotos

-0.38

logenetic

-0.37

مع

-0.36

 lección

-0.36

 frases

-0.35

 feedback

-0.35

escalier

-0.35

GetComponent

-0.34

 uovo

-0.34

子和

-0.34

POSITIVE LOGITS

GEBURTSDATUM

0.69

RegressionTest

0.65

 MainAxisSize

0.65

 ujednoznacz

0.60

VYMaps

0.60

ValueStyle

0.60

httphttps

0.57

новништво

0.53

 Réponses

0.52

SourceChecksum

0.51

Activations Density 0.069%

evidence and indicators of associations in various contexts

statements of fact or relationship

No Comments

No Known Activations