INDEX

Explanations

terms and functions related to programming, specifically involving lower and upper bounds in data structures

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

 للاسماء

-0.73

 surla

-0.69

 وتسجيلات

-0.62

tvguidetime

-0.61



-0.60

Aiheesta

-0.56

Ekster

-0.54

 parachoque

-0.54

期刊论文

-0.54

fillType

-0.53

POSITIVE LOGITS

HasIndex

0.43

search

0.40

bise

0.40

ureusement

0.36

 search

0.35

Vader

0.34

いかに

0.33

 Search

0.32

yargs

0.31

cold

0.30

Activations Density 0.006%

terms and functions related to programming, specifically involving lower and upper bounds in data structures

No Comments

No Known Activations