INDEX

Explanations

expressions of desire or intent involving the word "to."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 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

 adaptés

-0.39

 militares

-0.39

 evaluaciones

-0.38

 Millar

-0.37

 risked

-0.36

 risking

-0.36

 devastated

-0.36

 tatuagens

-0.36

 vegano

-0.36

 despite

-0.35

POSITIVE LOGITS

enumii

0.64

 يتيمه

0.64

 للمعارف

0.62



0.61

 Autorizaciones

0.61

rungsseite

0.60

principalColumn

0.58

 صوتيه

0.58

 betweenstory

0.57

NameInMap

0.57

Activations Density 0.081%

expressions of desire or intent involving the word "to."

No Comments

No Known Activations