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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

awtextra

-0.68

makeConstraints

-0.62

ftagPool

-0.62

 propOrder

-0.61

 ***!

-0.59

ніципалі

-0.59

UniformLocation

-0.58

 يتيمه

-0.57

fjspx

-0.57

endpush

-0.57

POSITIVE LOGITS

‐

0.34

search

0.34

style

0.33

fillColor

0.32

print

0.32

definecolor

0.32

page

0.32

 Brücke

0.31

ớp

0.31

Search

0.31

Activations Density 0.000%

No Known Activations

This feature has no known activations.