INDEX

Explanations

terms related to complementarity in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

Personendaten

-1.03

Smokey

-0.70

 Smoking

-0.69

smoke

-0.68

wax

-0.66

wikidata

-0.66

 Gown

-0.65

JTable

-0.65

Smoking

-0.64

 Rockefeller

-0.63

POSITIVE LOGITS

 supplement

1.76

 supplements

1.65

 complement

1.63

 supplementing

1.60

 complements

1.59

 Complement

1.51

 complément

1.50

 Supplement

1.48

Supplement

1.47

 SUPPLEMENT

1.46

Activations Density 0.187%

terms related to complementarity in various contexts

No Comments

No Known Activations