INDEX

Explanations

numerical citations or references to specific values in academic or technical documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

===============

-0.65

aktery

-0.65

dza

-0.63

مصادر

-0.63

ので

-0.63

istic

-0.63

vle

-0.62

bule

-0.61

 control

-0.61

vatel

-0.61

POSITIVE LOGITS

1.80

 NINE

1.19

nine

1.14

 Ninth

1.13

 ninth

1.05

 nueve

1.04

۹

1.03

ninth

1.02

 ninety

1.01

jestic

0.99

Activations Density 0.651%

numerical citations or references to specific values in academic or technical documents

No Comments

No Known Activations