INDEX

Explanations

technical terms and specific instances of the word "pan."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 Werke

-0.50

paedia

-0.47

Fiche

-0.40

мак

-0.40

 hObject

-0.40

Rode

-0.40

єм

-0.40

 gains

-0.39

setDo

-0.39

Inspiration

-0.39

POSITIVE LOGITS

pan

1.60

Pan

1.40

Pan

1.38

pan

1.31

 pans

1.25

PAN

1.23

PAN

1.22

pans

1.21

 Pans

1.04

 initComponents

0.85

Activations Density 1.995%

technical terms and specific instances of the word "pan."

No Comments

No Known Activations

technical terms and specific instances of the word "pan."

No Comments

No Known Activations