INDEX

Explanations

references to academic publishers or institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

Tikang

-0.54

 przec

-0.49

 surla

-0.47

 يتيمه

-0.46

Становништво

-0.44

TargetApi

-0.42

Jeografia

-0.42

arrings

-0.40

Биография

-0.40

şört

-0.40

POSITIVE LOGITS

 Publishers

0.48

 judiciales

0.46

ImageIO

0.42

 company

0.42

 Company

0.41

TableRow

0.40

 publishers

0.40

endcsname

0.40

 Verlags

0.39

念头

0.38

Activations Density 0.231%

references to academic publishers or institutions

No Comments

No Known Activations