INDEX

Explanations

references to villages and localities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kenya

-0.46

 Spector

-0.44

 Turquie

-0.43

 Heineken

-0.43

 Mexiko

-0.43

 Suède

-0.42

 Karlsruhe

-0.42

 Swanson

-0.41

ędzynarod

-0.41

 CANADIAN

-0.41

POSITIVE LOGITS

 village

1.12

 town

1.10

village

1.04

 villages

1.02

town

1.02

Village

0.99

 Village

0.96

 towns

0.95

Town

0.94

 Town

0.93

Activations Density 0.086%

references to villages and localities

No Comments

No Known Activations