INDEX

Explanations

content related to research studies and their findings, particularly focusing on published articles and scientific data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 importance

-0.40

 difficulty

-0.35

-0.33

 minimal

-0.33

 difficulties

-0.32

⇑

-0.32

 sort

-0.32

GeoNames

-0.32

 nice

-0.30

POSITIVE LOGITS

GEBURTSDATUM

0.74

 ribuan

0.62

arashtra

0.61

 AssemblyCompany

0.60

ulongan

0.56

Παραπομπές

0.56

audiovisuel

0.55

 lisäksi

0.54

ávajú

0.54

 MainAxisSize

0.53

Activations Density 0.715%

content related to research studies and their findings, particularly focusing on published articles and scientific data

No Comments

No Known Activations