INDEX

Explanations

numerical simplifications or evaluations

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

 فريبيس

-1.01

+#+#

-0.92

XtraReports

-0.86

 مشين

-0.83

principalTable

-0.79

ReusableCell

-0.76

 ويكيميديا

-0.75

󠁮

-0.75

 BoxFit

-0.73

 متعلقه

-0.72

POSITIVE LOGITS

 PUBLICATION

0.41

 branch

0.40

vr

0.40

hyrchwyd

0.40

 Branch

0.40

كين

0.39

lem

0.39

 fundamental

0.38

rmt

0.38

Cyfeiriadau

0.38

Activations Density 0.002%

numerical simplifications or evaluations

No Comments

No Known Activations