INDEX

Explanations

mathematical expressions and terminology related to graph theory and group theory

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

webElement

-0.62

+#+

-0.60

Hentet

-0.57

#+#

-0.56

SharedCtor

-0.56

 StringTokenizer

-0.55

 poffible

-0.55

 houſe

-0.54

EndInit

-0.54

 الحره

-0.54

POSITIVE LOGITS

 representations

0.42

 instituições

0.41

 respuestas

0.39

hadiran

0.38

 Stärke

0.38

 résister

0.38

 respostas

0.38

httphttps

0.38

ย

0.37

rów

0.37

Activations Density 0.620%

mathematical expressions and terminology related to graph theory and group theory

No Comments

No Known Activations