INDEX

Explanations

references to specific seasons and associated content in entertainment, particularly television

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

 تضيفلها

-0.71

 estekak

-0.69

ništ

-0.69

anmoins

-0.66

 WaitForSeconds

-0.65

 henvisninger

-0.62

TestTools

-0.59

 <=",

-0.59

 ModelExpression

-0.59

igshid

-0.57

POSITIVE LOGITS

зумі

0.54

 Baillargeon

0.54

 Zwick

0.53

setof

0.52

rippa

0.51

satunya

0.50

VOT

0.49

вань

0.49

 aggi

0.48

ByIndex

0.48

Activations Density 0.248%

references to specific seasons and associated content in entertainment, particularly television

No Comments

No Known Activations