INDEX

Explanations

references to user details and class object properties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

ViewFeatures

-0.62

Obrázky

-0.61

ocally

-0.55

المناصب

-0.54

ArrowToggle

-0.53

tanleria

-0.53

 intptr

-0.51

 utafitiHapana

-0.50

weit

-0.50

 appelez

-0.49

POSITIVE LOGITS

 RIPRODUZIONE

0.61

 mắn

0.59

strcat

0.58

$"{

0.54

 vôtre

0.53

`${

0.53

 joined

0.53

 Riproduzione

0.53

 greateſt

0.52

 nôtre

0.52

Activations Density 0.154%

references to user details and class object properties

No Comments

No Known Activations