INDEX

Explanations

references to programming or library functions related to matrix or numerical computations

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

帖最后由

-0.51

unknownFields

-0.44

TRAILING

-0.40

iprot

-0.40

richTextPanel

-0.38

 arşivlendi

-0.37

+#+

-0.36

\{\\

-0.36

scriptcase

-0.36

 समीक्षाओं

-0.36

POSITIVE LOGITS

AutoField

0.54

niająca

0.52

Vidite

0.50

 Organisateur

0.44

Citiți

0.43

的说道

0.43

 Juniors

0.43

contentLoaded

0.42

 Seniors

0.42

 Former

0.42

Activations Density 0.005%

references to programming or library functions related to matrix or numerical computations

No Comments

No Known Activations