INDEX

Explanations

functions or methods related to adding items or elements in programming contexts

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

readObject

-0.48

 Driscoll

-0.48

 Breen

-0.47

 Eder

-0.46

 AssemblyProduct

-0.45

 unheard

-0.44

getExternal

-0.44

дин

-0.43

bParam

-0.43

NegativeButton

-0.42

POSITIVE LOGITS

 Roskov

0.63

 defaultstate

0.61

 viewDidLoad

0.59

 nahilalakip

0.56

IContainer

0.55

kháu

0.46

ScopeManager

0.46

 kasarigan

0.45

 CURIAM

0.45

 незавершена

0.45

Activations Density 0.000%

functions or methods related to adding items or elements in programming contexts

No Comments

No Known Activations