INDEX

Explanations

structured data or code segments with hierarchical organization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 acceptez

-0.58

 هيا

-0.55

oredCriteria

-0.54

 unités

-0.53

 chimiques

-0.53

********/

-0.52

 couv

-0.52

 ujednoznacz

-0.51

 couvrir

-0.51

 modalidad

-0.51

POSITIVE LOGITS

 للمعارف

0.79

InputBorder

0.72

FormTagHelper

0.70

HtmlAttribute

0.64

 препратки

0.61

 AssemblyProduct

0.59

oader

0.59

WillAppear

0.59

SequentialGroup

0.58

colorbar

0.57

Activations Density 0.028%

structured data or code segments with hierarchical organization

No Comments

No Known Activations