INDEX

Explanations

references to user input or interactions

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

Legături

-0.33

 vacía

-0.33

 nahilalakip

-0.32

 publier

-0.32

aronder

-0.32

 TÉCN

-0.31

 dependiendo

-0.31

if

-0.30

 powied

-0.30

StructEnd

-0.30

POSITIVE LOGITS

 wireType

0.70

 ChatGPT

0.67

#+#

0.58

 ſte

0.56

 faſt

0.54

 Autorisations

0.53

 purpoſe

0.51

 Anſ

0.51

astify

0.49

 ſind

0.49

Activations Density 1.703%

references to user input or interactions

No Comments

No Known Activations