INDEX

Explanations

active verbs and phrases related to daily tasks and processes

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 personally

-0.58

 personal

-0.49

 Huns

-0.47

shima

-0.42

ucci

-0.41

pel

-0.41

 Pelo

-0.41

 persoon

-0.40

 espoir

-0.40

burg

-0.39

POSITIVE LOGITS

 itself

1.08

 Itself

1.05

itself

0.97

its

0.91

SharedCtor

0.83

Its

0.82

Its

0.77

 صوتيه

0.76

ValueStyle

0.74

 autorytatywna

0.74

Activations Density 1.037%

active verbs and phrases related to daily tasks and processes

No Comments

No Known Activations