INDEX

Explanations

references to "strides" in programming contexts

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

 Popp

-0.44

Pen

-0.42

 Popular

-0.41

Hel

-0.41

 Administrativna

-0.40

 Púb

-0.38

 Paul

-0.38

Bon

-0.38

Pap

-0.38

Popular

-0.37

POSITIVE LOGITS

 stride

1.19

stride

1.16

Stride

1.10

strides

0.90

 strides

0.80

发表于

0.63

ChildScrollView

0.62

SequentialGroup

0.62

matchCondition

0.59

JvmStatic

0.58

Activations Density 0.002%

references to "strides" in programming contexts

No Comments

No Known Activations