INDEX

Explanations

numeric values or specific data identifiers 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 @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 commerciali

-0.47

 aislados

-0.45

 élevées

-0.45

 épais

-0.44

 élevés

-0.43

僚

-0.43

 soutenir

-0.43

 lontano

-0.43

 melaporkan

-0.42

 nuages

-0.42

POSITIVE LOGITS

 للمعارف

1.27

+#+#

1.13

ⓧ

0.89

MigrationBuilder

0.85

########.

0.84

?>/

0.80

Спољашње

0.77

SBATCH

0.75

出版年

0.74

($)

0.72

Activations Density 0.471%

numeric values or specific data identifiers in a structured format

No Comments

No Known Activations