INDEX

Explanations

expressions related to attempts and intentions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

chartInstance

-0.56

 للمعارف

-0.55

HasAnnotation

-0.46

 noDo

-0.44

styleType

-0.44

Tembelea

-0.42

Datuak

-0.42

caire

-0.42

ostock

-0.41

 alcanza

-0.41

POSITIVE LOGITS

 planned

0.45

copg

0.44

JNIEnv

0.42

planned

0.38

 papp

0.36

enterOuterAlt

0.36

Planned

0.35

 voorbere

0.34

styleable

0.34

 но

0.34

Activations Density 0.669%

expressions related to attempts and intentions

No Comments

No Known Activations