INDEX

Explanations

boolean values, particularly those set to true or false

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

-0.50

())),

-0.50

"}},

-0.50

()?.

-0.49

-0.47

="@

-0.46

=";

-0.46

intios

-0.46

"},

-0.45

POSITIVE LOGITS

fjspx

0.49

enough

0.48

tagHelperRunner

0.48

SequentialGroup

0.48

GEBURTSDATUM

0.47

IntoConstraints

0.46

 enough

0.43

 viewDidLoad

0.42

 estekak

0.41

 Enough

0.40

Activations Density 0.193%

boolean values, particularly those set to true or false

No Comments

No Known Activations