INDEX

Explanations

phrases that indicate relationships or comparisons, particularly involving balance and conflict

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gridad

-0.41

着手

-0.39

 CommonModule

-0.36

 الرياضيه

-0.35

⅗

-0.35

 others

-0.33

 saldo

-0.33

others

-0.32

kün

-0.32

CopyWith

-0.31

POSITIVE LOGITS

 himo

0.51

jspb

0.49

 kaarangay

0.49

BASELINE

0.48

тьяна

0.48

 Мексичка

0.47

LEGGI

0.46

كويكب

0.45

ьаж

0.45

 Standard

0.45

Activations Density 0.816%

phrases that indicate relationships or comparisons, particularly involving balance and conflict

No Comments

No Known Activations