INDEX

Explanations

code-related keywords like "string", "scanner", or "optional"

oai_token-act-pair · gemini-2.0-flash

Code snippets

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

IGA

-0.43

dart

-0.41

ണ്

-0.40

xin

-0.40

="

-0.39

nino

-0.39

zielle

-0.39

eles

-0.38

lē

-0.38

tud

-0.38

POSITIVE LOGITS

 محفوظة

0.77

 للمعارف

0.77

 Jefus

0.75

 myſelf

0.75

 becauſe

0.74

ſelves

0.74

 itſelf

0.74

Efq

0.73

IntoConstraints

0.73

wiſe

0.71

Activations Density 3.026%

code-related keywords like "string", "scanner", or "optional"

Code snippets

No Known Activations