INDEX

Explanations

variations of the word "pe."

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

 Pelleg

-0.48

 oglą

-0.46

 الاع

-0.44

 Schlegel

-0.43

 Atem

-0.42

 ketat

-0.40

 Cuevas

-0.39

 Penge

-0.39

 Bowen

-0.39

immel

-0.38

POSITIVE LOGITS

pe

1.44

Pe

1.17

Pe

1.02

pe

0.85

PE

0.64

ces

0.57

UserScript

0.57

titleMargin

0.55

TypedDataSet

0.54

PE

0.52

Activations Density 0.172%

variations of the word "pe."

No Comments

No Known Activations

variations of the word "pe."

No Comments

No Known Activations