INDEX

Explanations

beginning and structural markers in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

onOptions

-0.65

rativa

-0.64

aarrggbb

-0.63

quelize

-0.63

ългария

-0.61

idoscope

-0.60

)";

-0.59

tangentMode

-0.59

GeneratedValue

-0.58

μισ

-0.58

POSITIVE LOGITS

 Roskov

0.63

 surla

0.62

 Hopf

0.59

fjspx

0.59

ToScroll

0.58

 birra

0.55

تقاوى

0.55

 StatelessWidget

0.54

 tarvit

0.54

DebuggerNonUser

0.53

Activations Density 0.013%

beginning and structural markers in the text

No Comments

No Known Activations