INDEX

Explanations

references to characters or elements associated with a specific fictional universe or narrative structure

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 AppCompatTheme

-0.69

 kasarigan

-0.63

führende

-0.53

amom

-0.53

($__

-0.52

thol

-0.52

haf

-0.52

colat

-0.51

ням

-0.51

orti

-0.51

POSITIVE LOGITS

 صوتيه

0.77

Hentet

0.65

DeleteBehavior

0.54

StructEnd

0.53

 виправивши

0.51

 SqlParameter

0.50

DoubleQuotes

0.46

Controllo

0.46

 مواليد

0.45

 Meksiku

0.45

Activations Density 1.550%

references to characters or elements associated with a specific fictional universe or narrative structure

No Comments

No Known Activations