INDEX

Explanations

No Explanations Found

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

matchCondition

-0.72

Personendaten

-0.62

TagMode

-0.60

httphttps

-0.59

ValueStyle

-0.53

 umě

-0.49

 صوتيه

-0.47

richTextPanel

-0.47

 мәгълү

-0.46

itisation

-0.46

POSITIVE LOGITS

TypedDataSet

0.41

findpost

0.41

MessageTagHelper

0.38

ectoria

0.36

spyOn

0.34

 Neat

0.34

 Meksiku

0.34

 diagonal

0.33

defineProperty

0.33

 reloadData

0.33

Activations Density 0.000%

No Known Activations

This feature has no known activations.