INDEX

Explanations

references to programming constructs and their functionality

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ँ

-0.50

resave

-0.44

Rüyada

-0.41

 PROGRAMME

-0.41

behaviour

-0.41

 GenerationType

-0.41

 formulae

-0.41

 diarrhoea

-0.40

rated

-0.40

 programme

-0.40

POSITIVE LOGITS

EndContext

0.55

 noDo

0.48

=${

0.46

=");

0.44

 समीक्षाएं

0.42

0.40

 $(\%)$

0.40

$=$

0.40

:");

0.38

Activations Density 0.097%

references to programming constructs and their functionality

No Known Activations