INDEX

Explanations

various forms and derivatives of the word "te."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

انجليز

-0.41

RegressionTest

-0.40

InlineData

-0.40

 mía

-0.40

//

-0.39

 mergeFrom

-0.38

 encantó

-0.37

IVEREF

-0.37

ArrowToggle

-0.36

 masiva

-0.36

POSITIVE LOGITS

Te

0.68

Tef

0.58

TE

0.58

Te

0.58

 Те

0.54

TET

0.54

AndEndTag

0.52

 Teas

0.50

Билгалдахарш

0.48

Те

0.47

Activations Density 0.448%

various forms and derivatives of the word "te."

No Comments

No Known Activations