INDEX

Explanations

punctuation marks and formatting symbols related to legal and patent references

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

 تضيفلها

-0.72

 bezeichneter

-0.62

JspWriter

-0.61

oredCriteria

-0.60

NameInMap

-0.60

ThroughAttribute

-0.59

 CanadaChoose

-0.57

webElementXpaths

-0.57

sizeCache

-0.57

featureID

-0.56

POSITIVE LOGITS

 wiki

0.41

 Trasp

0.41

wende

0.41

Configuration

0.40

née

0.40

 Configuration

0.40

Matrices

0.39

 passed

0.39

 missed

0.38

getLong

0.37

Activations Density 0.021%

punctuation marks and formatting symbols related to legal and patent references

No Comments

No Known Activations