INDEX

Explanations

programming constructs related to data structures and map entries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

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

:✨

-0.63

SharedDtor

-0.54

 kaarangay

-0.52

liothèque

-0.50

RegressionTest

-0.48

րջ

-0.47

Datuak

-0.47

ErrUnexpectedEOF

-0.47

 صوتيه

-0.47

RTLR

-0.47

POSITIVE LOGITS

withIdentifier

0.58

EndContext

0.56

 Brava

0.56

PerformLayout

0.54

 препратки

0.54

Obrázky

0.54

 cables

0.53

DVB

0.53

 Lauter

0.52

LNB

0.52

Activations Density 0.092%

programming constructs related to data structures and map entries

No Comments

No Known Activations