INDEX

Explanations

technical specifications and metrics in a structured format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

rungsseite

-0.63

 المعيارى

-0.50

 unknownFields

-0.50

 leggi

-0.48

\{\\

-0.47

Datuak

-0.46

 dialami

-0.46

iempo

-0.45

onChanged

-0.44

 близь

-0.44

POSITIVE LOGITS

 weighting

0.85

 kasarigan

0.75

Portály

0.67

 weighted

0.67

 weigh

0.65

weigh

0.64

Portale

0.64

 CURIAM

0.62

StructEnd

0.62

 weighed

0.62

Activations Density 0.302%

technical specifications and metrics in a structured format

No Comments

No Known Activations