INDEX

Explanations

terms and concepts related to electrical engineering and signal processing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

tagHelperRunner

-0.68

featureID

-0.61

CppMethod

-0.58

ReusableCell

-0.57

WithIOException

-0.56

 Meksiku

-0.54

UITableViewCell

-0.54

 ProtoMessage

-0.52

 autorytatywna

-0.52

delwed

-0.50

POSITIVE LOGITS

 reverse

0.58

 reversed

0.54

 inverted

0.52

 Reverse

0.51

 reverses

0.50

Reverse

0.49

 reversing

0.48

reverse

0.47

 opposite

0.46

 negative

0.44

Activations Density 1.046%

terms and concepts related to electrical engineering and signal processing

No Comments

No Known Activations