INDEX

Explanations

references to operators and their usage in programming contexts

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

INCREF

-0.39

<bos>

-0.38

 noDo

-0.37

 Flach

-0.36

 padx

-0.36

 familières

-0.35

 específica

-0.35

 uLocal

-0.35

 erkek

-0.35

 clothing

-0.34

POSITIVE LOGITS

Maps

0.76

 Maps

0.76

aux

0.75

maps

0.75

unknownFields

0.70

footer

0.69

Footer

0.68

protoc

0.62

navbar

0.62

Raj

0.60

Activations Density 0.159%

references to operators and their usage in programming contexts

No Comments

No Known Activations