INDEX

Explanations

technical terms and programming language syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

LETT

-0.56

 Godfrey

-0.55

 overloaded

-0.50

 Boyce

-0.50

lė

-0.50

IGE

-0.49

essica

-0.47

 Hector

-0.47

</tbody>

-0.47

Sac

-0.46

POSITIVE LOGITS

bin

0.89

sin

0.87

hin

0.86

lin

0.85

gin

0.85

min

0.84

cin

0.84

in

0.82

fin

0.81

din

0.80

Activations Density 0.338%

technical terms and programming language syntax

No Comments

No Known Activations