INDEX

Explanations

language-related terms or keywords

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

__':

-0.83

MigrationBuilder

-0.82

SourceChecksum

-0.81

 estekak

-0.76

 typelib

-0.76

}');

-0.73

 Thayer

-0.70

__":

-0.69

reactivex

-0.69

ChildScrollView

-0.68

POSITIVE LOGITS

 Lang

0.64

 LANG

0.53

Lang

0.51

lang

0.50

 Langley

0.49

 lang

0.49

reflect

0.47

UNIX

0.46

String

0.46

laient

0.43

Activations Density 0.038%

language-related terms or keywords

No Comments

No Known Activations