INDEX

Explanations

technical terms and quantifiable metrics in scientific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 unut

-0.51

 niin

-0.42

!***

-0.37

אז

-0.37

 hồn

-0.37

↵↵

-0.37

ג

-0.36

mu

-0.36

afar

-0.35

 anton

-0.35

POSITIVE LOGITS

évaluateur

0.96

 المعيارى

0.96

 فريبيس

0.93

 Мексичка

0.88

ViewFeatures

0.85

monton

0.79

 invokingState

0.79

aarrggbb

0.79

haustible

0.75

 gawas

0.75

Activations Density 2.080%

technical terms and quantifiable metrics in scientific contexts

No Comments

No Known Activations