INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 Phry

-1.03

 itſelf

-1.00

 Mahomet

-0.99

 Houſe

-0.98

 Huguen

-0.97

 myſelf

-0.91

 Shakspeare

-0.91

 doubtnut

-0.90

 raiſ

-0.89

 Agamemnon

-0.88

POSITIVE LOGITS

the

1.56

 same

1.16

The

1.11

THE

1.05

The

0.98

 latter

0.96

 entire

0.95

new

0.94

 final

0.93

 most

0.91

Activations Density 0.000%

No Known Activations

This feature has no known activations.