INDEX

Explanations

comparisons and contrasts between data points or study results

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

Connell

-0.40

copg

-0.35

 tambi

-0.35

 которую

-0.35

 fenêtres

-0.32

 podían

-0.32

 Seorang

-0.31

 które

-0.31

 también

-0.31

zzleHttp

-0.30

POSITIVE LOGITS

GEBURTSDATUM

0.88

 kaarangay

0.81

migrationBuilder

0.77

rungsseite

0.73

FieldNumber

0.67

 tartalomajánló

0.67

 חיצוניים

0.67

 مشين

0.65

 CascadeType

0.65

Cyfeiriadau

0.64

Activations Density 1.060%

comparisons and contrasts between data points or study results

No Comments

No Known Activations