INDEX

Explanations

words referring to abstract concepts, often negative, and sometimes related to computing or medicine

oai_token-act-pair · gemini-2.0-flash

discussion, announcements

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_7/width_16k/average_l0_70

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

脚注の使い方

-0.79

httphttps

-0.71

 lenker

-0.68

 BoxDecoration

-0.68

adə

-0.66

Manbalar

-0.65

bootstrapcdn

-0.65

ẢN

-0.64

 OkHttpClient

-0.63

complexContent

-0.63

POSITIVE LOGITS

<bos>

0.77

 AssemblyVersion

0.58

Nationalité

0.47

0.46

...@

0.46

 colspan

0.46

Geplaatst

0.44

جستارهای

0.44

 đông

0.43

HostException

0.42

Activations Density 38.676%

words referring to abstract concepts, often negative, and sometimes related to computing or medicine

discussion, announcements

No Comments

No Known Activations