INDEX

Explanations

the start of a document or significant section marked by a specific token

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

Non-English language fragments

np_acts-logits-general · gemini-2.0-flash

Chinese, Japanese, Korean, and Russian

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

 الحره

-1.02

tvguidetime

-1.02

 تانيه

-0.84

ArgsConstructor

-0.82

 various

-0.82

findpost

-0.78

 للاسماء

-0.76

=$?

-0.71

AutoScaleMode

-0.71

 resourceCulture

-0.71

POSITIVE LOGITS

باره

0.46

 mnie

0.41

veo

0.41

 lagi

0.40

nější

0.40

 μας

0.39

 dAtA

0.39

 ترین

0.39

bný

0.39

 σας

0.39

Activations Density 0.046%

the start of a document or significant section marked by a specific token

Non-English language fragments

Chinese, Japanese, Korean, and Russian

No Comments

No Known Activations