INDEX

Explanations

references to NIL and optional values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 autorytatywna

-0.49

 PeEnEo

-0.45

BufferException

-0.45

batore

-0.43

AnimationsModule

-0.42

 ángulo

-0.41

 resourceCulture

-0.40

 castelo

-0.39

ValueStyle

-0.38

 campamento

-0.38

POSITIVE LOGITS

 native

0.63

 desktop

0.60

期刊论文

0.56

ks

0.51

WriteTagHelper

0.50

خواندن

0.50

 DESK

0.50

 Desktop

0.50

native

0.50

 expecting

0.49

Activations Density 0.206%

references to NIL and optional values

No Comments

No Known Activations