INDEX

Explanations

strings that denote data structures or programming constructs in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 poffible

-0.97

 myſelf

-0.95

Geplaatst

-0.94

évaluateur

-0.91

 leaſt

-0.91

 initComponents

-0.88

 juſt

-0.84

ChildScrollView

-0.83

 occaf

-0.81

 leek

-0.81

POSITIVE LOGITS

 strings

1.54

 string

1.48

 Stringer

1.42

strings

1.38

STRING

1.35

 Strings

1.29

 STRING

1.29

string

1.27

Strings

1.26

 String

1.24

Activations Density 0.094%

strings that denote data structures or programming constructs in code

No Comments

No Known Activations