INDEX

Explanations

code structures and indicators related to programming, specifically annotations and class definitions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

medriver

-0.79

RegressionTest

-0.64

 >=",

-0.62

ibouti

-0.62

 vulgares

-0.61

 Optic

-0.61

 Rogan

-0.56

 buddha

-0.55

ulkner

-0.55

 Vole

-0.55

POSITIVE LOGITS

Abp

0.63

 كومونز

0.62

 IBOutlet

0.61

(&:

0.60

endpush

0.59

IBOutlet

0.59

Datuak

0.57

HandleFunc

0.57

Portail

0.56

IBAction

0.55

Activations Density 0.062%

code structures and indicators related to programming, specifically annotations and class definitions

No Comments

No Known Activations