INDEX

Explanations

terms and phrases related to assessments, evaluations, and contexts of importance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 warszawa

-0.65

 houſe

-0.62

resents

-0.61

LayoutConstraint

-0.58

 ſch

-0.56

 собою

-0.55

 niedersachsen

-0.55

 themſelves

-0.55

 pleaſure

-0.54

évaluateur

-0.54

POSITIVE LOGITS

of

0.71

 estekak

0.71

 duquel

0.63

 تضيفلها

0.57

 CreateTagHelper

0.56

 whereof

0.55

__*/

0.54

elemField

0.53

 الرياضيه

0.53

 resourceCulture

0.52

Activations Density 0.371%

terms and phrases related to assessments, evaluations, and contexts of importance

No Comments

No Known Activations