INDEX

Explanations

terms related to reduction or minimization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

]!='

-0.67

 CURIAM

-0.66

 Wikispecies

-0.64

psia

-0.63

cupa

-0.63

エム

-0.61

odils

-0.61

Appellants

-0.60

askar

-0.60

 Jank

-0.59

POSITIVE LOGITS

 reduc

1.37

 Reduce

1.36

Reduce

1.29

 reduce

1.29

 reduction

1.27

 Reduced

1.25

reduce

1.25

 REDUC

1.24

 reducers

1.24

 reducing

1.22

Activations Density 0.096%

terms related to reduction or minimization

No Comments

No Known Activations