INDEX

Explanations

references to specific units of measurement or containers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

:+:

-0.52

 viewHolder

-0.47

نان

-0.47

casi

-0.46

 Bruno

-0.46

corn

-0.46

igen

-0.45

AndEndTag

-0.45

 newName

-0.45

rington

-0.44

POSITIVE LOGITS

ISupport

0.82

 informée

0.82

SBATCH

0.80

 otomatig

0.80

 myſelf

0.79

basket

0.79

 dumps

0.78

 ſtand

0.77

 ApiException

0.77

.}~\

0.77

Activations Density 0.070%

references to specific units of measurement or containers

No Comments

No Known Activations