INDEX

Explanations

phrases related to personal preferences and opinions

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

 arşivlendi

-0.45

IContainer

-0.42

AnchorTagHelper

-0.37

mergeFrom

-0.35

AutoScaleMode

-0.35

EndProject

-0.35

KommentareTeilen

-0.35

Przypisy

-0.35

 thefts

-0.34

zeera

-0.34

POSITIVE LOGITS

uxxxx

0.73

 Italijani

0.57

хьтан

0.48

tvguidetime

0.48

 payday

0.47

 Wikimedijinoj

0.44

цездатний

0.43

 Roskov

0.42

PageSize

0.41

 kasarigan

0.41

Activations Density 0.077%

phrases related to personal preferences and opinions

No Comments

No Known Activations