INDEX

Explanations

numerical identifiers and references related to academic papers or articles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

Personendaten

-0.87

 tartalomajánló

-0.84

parsedMessage

-0.84

=$?

-0.83

UnusedPrivate

-0.80

SharedCtor

-0.80

featureID

-0.76

 дописавши

-0.76

">—

-0.75

IBarButtonItem

-0.74

POSITIVE LOGITS

<bos>

0.49

Források

0.48

 فريبيس

0.45

chtet

0.43

følgelig

0.39

ेशा

0.39

__);

0.38

ódó

0.37

ableness

0.36

Activations Density 0.267%

numerical identifiers and references related to academic papers or articles

No Comments

No Known Activations