INDEX

Explanations

classes and objects related to programming and 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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 Zustimmung

-0.39

subequations

-0.36

astype

-0.34

 LIFT

-0.34

Todavía

-0.34

 consejos

-0.34

katapos

-0.33

getColumnIndex

-0.33

toThrow

-0.33

 transfieras

-0.33

POSITIVE LOGITS

ArgsConstructor

0.75

новништво

0.58

AutoScaleMode

0.53

InjectAttribute

0.53

 NgModule

0.50

Enllaces

0.50



0.49

 الحره

0.48

parsedMessage

0.48

BeginInit

0.46

Activations Density 0.020%

classes and objects related to programming and data structures

No Comments

No Known Activations