INDEX

Explanations

specific sequences of letters and symbols often found in code or technical documentation

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

bollah

-0.37

mCurrent

-0.35

いけない

-0.35

runOn

-0.34

Dimanche

-0.33

Bedrooms

-0.33

 capilar

-0.32

Lundi

-0.32

 Фин

-0.32

POSITIVE LOGITS

ImageContext

0.69

ade

0.63

ChildScrollView

0.60

de

0.60

ce

0.60

te

0.59

ame

0.58

ile

0.58

ide

0.58

Географиясе

0.58

Activations Density 0.850%

specific sequences of letters and symbols often found in code or technical documentation

No Comments

No Known Activations