INDEX

Explanations

patterns related to numerical data or coding constructs

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

+#+#

-0.78

OGND

-0.68

RegressionTest

-0.67

GraphicsUnit

-0.65

 Infórmanos

-0.59

bootstrapcdn

-0.58

TagMode

-0.57

\{\\

-0.57

 useAppContext

-0.56

 Roskov

-0.56

POSITIVE LOGITS

حوالہ

0.40

 hơi

0.35

日

0.34

之前的

0.32

 físicas

0.32

 şeyler

0.32

 robić

0.32

 légèrement

0.31

 νό

0.31

 tänka

0.31

Activations Density 0.006%

patterns related to numerical data or coding constructs

No Comments

No Known Activations