INDEX

Explanations

references to Newton and related concepts

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

وردار

-0.52

 Soriano

-0.49

)['

-0.47

 OnEnable

-0.46

abbix

-0.45

ⓘ

-0.45

 Jorg

-0.44

 Doran

-0.42

Joe

-0.42

JOE

-0.42

POSITIVE LOGITS

 Newton

1.43

Newton

1.34

 newton

1.24

 NEWTON

1.19

newton

1.17

 Newtonian

0.92

 Newtown

0.75

 Newt

0.67

 feroit

0.66

MemoryWarning

0.66

Activations Density 0.007%

references to Newton and related concepts

No Comments

No Known Activations