INDEX

Explanations

references to test cases and testing-related terms

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

Code testing related filenames

np_acts-logits-general · gemini-2.0-flash

testing code, suites, utilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

ftagPool

-0.62

parsedMessage

-0.60

 kaynağından

-0.59

 ModelExpression

-0.56

AndEndTag

-0.56

EDEFAULT

-0.56

 GoogleFonts

-0.55

writeFieldEnd

-0.55

 Superan

-0.54

 Мексичка

-0.53

POSITIVE LOGITS

utils

0.46

 Suites

0.45

 suite

0.40

 purposes

0.40

junit

0.38

tools

0.37

Helper

0.36

helper

0.36

 Suite

0.36

suite

0.35

Activations Density 0.093%

references to test cases and testing-related terms

Code testing related filenames

testing code, suites, utilities

No Comments

No Known Activations