INDEX

Explanations

terms related to negation or differentiation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 препратки

-0.99

 pleins

-0.91

 AssemblyCulture

-0.88

fromCharCode

-0.79

 plein

-0.79

duire

-0.78

tableFuture

-0.77

펴

-0.77

 calendriers

-0.76

cetin

-0.76

POSITIVE LOGITS

DIS

1.40

Dis

1.32

DIS

1.29

Dis

1.28

 useDispatch

1.20

dis

1.17

dis

1.08

 Disqus

0.99

 dispens

0.94

 Driscoll

0.89

Activations Density 0.177%

terms related to negation or differentiation

No Comments

No Known Activations