INDEX

Explanations

phrases related to features and specifications of products

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

ⓧ

-0.74

saraba

-0.73

 CreateTagHelper

-0.63

 StringTokenizer

-0.60

DeleteBehavior

-0.60

 disambiguazione

-0.59

 Савезне

-0.59

 saites

-0.58

=$?

-0.57

iseerd

-0.57

POSITIVE LOGITS

 préférences

0.56

urich

0.50

Util

0.48

('/:

0.45

 titulada

0.44

?>"

0.44

 Conceptual

0.44

loyds

0.43

hline

0.43

 Util

0.43

Activations Density 0.149%

phrases related to features and specifications of products

No Comments

No Known Activations