INDEX

Explanations

sequences of numeric values and formatting in code snippets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

Personensuche

-0.83

 kasarigan

-0.76

 فريبيس

-0.70

 kaynağından

-0.69

 MatDialog

-0.69

<>());

-0.68

➢

-0.66

 Spiro

-0.65

 Touren

-0.63

éb

-0.63

POSITIVE LOGITS

0.55

equalsIgnoreCase

0.55

PWN

0.54

Eug

0.54

Efq

0.54

0.53

 ditto

0.53

 Zinn

0.53

ısını

0.52

Hemp

0.52

Activations Density 0.090%

sequences of numeric values and formatting in code snippets

No Comments

No Known Activations