INDEX

Explanations

details related to processes and outcomes in specific 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-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

ChildScrollView

-0.61

IActionResult

-0.54

лтемелер

-0.51

 OMITBAD

-0.49

MessageOf

-0.49

 TextAlign

-0.49

 հղումներ

-0.46

הערות

-0.44

openhauer

-0.42

onOptions

-0.41

POSITIVE LOGITS

ſp

0.48

 ſtate

0.48

 ſte

0.47

 houſe

0.46

 Houſe

0.46

Diweddarwch

0.44

 ſta

0.44

 faſt

0.43

 vectorielles

0.43

 Diſ

0.43

Activations Density 1.205%

details related to processes and outcomes in specific contexts

No Comments

No Known Activations