INDEX

Explanations

instances of numbers or values related to statistics and data

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 gärna

-0.47

 insiste

-0.45

 savent

-0.45

 esternos

-0.44

पया

-0.43

 noDo

-0.43

 dipasang

-0.43

 überrascht

-0.43

 illustratie

-0.42

Rüyada

-0.42

POSITIVE LOGITS

 thousand

0.59

 thousands

0.58

0.55

 THOUS

0.54

 millions

0.52

 poison

0.48

thousands

0.47

 tens

0.47

#,

0.47

,,

0.47

Activations Density 0.207%

instances of numbers or values related to statistics and data

No Comments

No Known Activations