INDEX

Explanations

terms related to normalcy and normalization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

scribers

-0.40

 onPostExecute

-0.38

 lentejuelas

-0.36

tagHelperRunner

-0.34

äischen

-0.34

 kır

-0.32

easy

-0.32

ố

-0.32

 vorliegenden

-0.32

ész

-0.32

POSITIVE LOGITS

 Normal

0.98

Normal

0.93

 NORMAL

0.93

NORMAL

0.87

normal

0.81

 normal

0.79

 normales

0.73

 Norma

0.71

 المعيارى

0.69

 normale

0.69

Activations Density 0.222%

terms related to normalcy and normalization

No Comments

No Known Activations