INDEX

Explanations

mentions of ordinal numbers and references to awards

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

er

-0.69

 either

-0.58

aluz

-0.57

OrderDetail

-0.56

 оригіналу

-0.52

 Hutchins

-0.52

erden

-0.51

šev

-0.51

teek

-0.51

hintText

-0.51

POSITIVE LOGITS

 tenth

1.34

 ninth

1.34

tenth

1.33

 seventh

1.25

 sixth

1.25

 twelfth

1.24

 eleventh

1.23

Fifth

1.23

 Tenth

1.23

 eighth

1.22

Activations Density 0.132%

mentions of ordinal numbers and references to awards

No Comments

No Known Activations