INDEX

Explanations

references to documentation and systems errors

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

linking words to descriptions

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Chriftian

-0.57

таратура

-0.49

 miniaturka

-0.46

 deſt

-0.40

 desmotivaciones

-0.40

 aveug

-0.38

colgante

-0.37

 fubject

-0.35

 fhew

-0.35

 Diſ

-0.35

POSITIVE LOGITS

uxxxx

0.60

expandindo

0.60

verifyException

0.57

MigrationBuilder

0.55

፩

0.48

पया

0.48

archiviato

0.43

 tartalomajánló

0.43

 ويكيپيديا

0.43

oneofs

0.43

Activations Density 0.069%

references to documentation and systems errors

linking words to descriptions

No Comments

No Known Activations