INDEX

Explanations

symbols and identifiers related to programming or coding syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

expandindo

-0.47

assertIn

-0.42

MockBean

-0.40

TagHelper

-0.39

 مشين

-0.37

Чыганаклар

-0.37

 MongoClient

-0.37

/*

-0.35

 كومونز

-0.35

WritableDatabase

-0.35

POSITIVE LOGITS

ReusableCell

0.51

AndEndTag

0.50

 vastaan

0.49

 Früchte

0.48

 arvio

0.46

richTextPanel

0.46

MLLoader

0.44

Encyklopedia

0.44

urlencoded

0.43

zbęd

0.43

Activations Density 0.001%

symbols and identifiers related to programming or coding syntax

No Comments

No Known Activations