INDEX

Explanations

references to time duration and experiences over the years

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

 Wharton

-0.36

 mahd

-0.34

 atsi

-0.34

 reflexes

-0.34

 voies

-0.34

 vone

-0.33

 napa

-0.32

OCA

-0.32

ity

-0.32

cl

-0.32

POSITIVE LOGITS

ずっと

0.60

 Monfieur

0.59

前から

0.59

 يتيمه

0.56

 Anſ

0.55

 faſt

0.54

一直

0.53

λευτα

0.52

tagHelperRunner

0.50

setVerticalGroup

0.50

Activations Density 0.029%

references to time duration and experiences over the years

No Comments

No Known Activations