INDEX

Explanations

specific phrases indicating locations and positions

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

 насељу

-0.64

 referenties

-0.63

后面的

-0.50

 afield

-0.49

مصادر

-0.49

 kuchni

-0.47

ligg

-0.46

indeki

-0.45

 Belakang

-0.45

 ventr

-0.44

POSITIVE LOGITS

 typelib

0.84

errHandler

0.70

sizeCache

0.67

sendRedirect

0.67

WebServlet

0.65

windowFixed

0.64

CppMethod

0.63

 LoggerFactory

0.62

AndEndTag

0.62

InvalidProtocol

0.62

Activations Density 0.204%

specific phrases indicating locations and positions

No Comments

No Known Activations