INDEX

Explanations

numerical identifiers and references within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

canestro

-0.37

ContentAlignment

-0.33

AndUpdate

-0.32

XtraGrid

-0.32

mathematical

-0.31

駅

-0.30

arande

-0.28

bentukan

-0.28

necessary

-0.28

駅

-0.28

POSITIVE LOGITS

 stessi

0.60

 dezelfde

0.56

 hetzelfde

0.54

 stej

0.54

 gleicher

0.53

Хьажоргаш

0.53

 same

0.53

同じく

0.52

 precedenti

0.52

ckså

0.51

Activations Density 2.760%

numerical identifiers and references within the text

No Comments

No Known Activations