INDEX

Explanations

statistical terms related to effectiveness and reliability measurements in research

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

 otomatig

-0.48

 Савезне

-0.44

GEBURTS

-0.39

ukone

-0.39

 aufs

-0.38

ensory

-0.38

matchCondition

-0.37

wapV

-0.36

 exped

-0.36

Exped

-0.36

POSITIVE LOGITS

instancetype

0.41

Rohy

0.40

ApiModelProperty

0.39

 Tabor

0.39

GeneratedMessage

0.39

 NDEBUG

0.39

eddable

0.39

 Mubarak

0.39

FailureListener

0.39

migrationBuilder

0.37

Activations Density 0.163%

statistical terms related to effectiveness and reliability measurements in research

No Comments

No Known Activations