INDEX

Explanations

technical terminology related to statistical methods and modeling

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

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

+:+

-0.67

 wireType

-0.59

 tangan

-0.54

 للاسماء

-0.53

ي

-0.52

Abitanti

-0.51

י

-0.50

็จ

-0.50

️

-0.48

POSITIVE LOGITS

 eccl

0.68

henge

0.68

AndEndTag

0.67

LEIA

0.67

toHaveBeenCalled

0.66

CWE

0.63

 наконец

0.63

 pady

0.63

 prostu

0.63

Activations Density 0.645%

technical terminology related to statistical methods and modeling

No Comments

No Known Activations