INDEX

Explanations

phrases indicating benefits and appreciation in various contexts

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

 تضيفلها

-0.45

httphttps

-0.42

 незавершена

-0.41

AdapterView

-0.40

CopyWith

-0.39

 comuniques

-0.37

 MessageBoxIcon

-0.37

ramientas

-0.36

Chham

-0.36

 Mono

-0.35

POSITIVE LOGITS

 Aufg

0.45

 المث

0.44

Спо

0.44

 spontaneous

0.43

NameInMap

0.42

posedge

0.39

دانشنامهٔ

0.39

 einges

0.39

arusian

0.39

NSON

0.38

Activations Density 0.018%

phrases indicating benefits and appreciation in various contexts

No Comments

No Known Activations