INDEX

Explanations

references to reports and studies in discussions

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

 Câmara

-0.35

manfaat

-0.32

ardless

-0.32

farwyddwr

-0.32

 pela

-0.32

 physician

-0.31

 government

-0.30

diteur

-0.30

kot

-0.30

adə

-0.30

POSITIVE LOGITS

 للاسماء

0.61

addCriterion

0.58

 ModelExpression

0.57

AddTagHelper

0.55

 protoimpl

0.53

XmlAccessType

0.52

bootstrapcdn

0.52

"]}

0.52

FunctionFlags

0.50

 ویکی‌پدی

0.50

Activations Density 0.122%

references to reports and studies in discussions

No Comments

No Known Activations