INDEX

Explanations

references to common concepts or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 GenerationType

-0.62

pperware

-0.56

TypedDataSet

-0.55

iermark

-0.52

igurumi

-0.52

MLLoader

-0.52

 onPostExecute

-0.52

 endblock

-0.51

ValueStyle

-0.51

 MaterialApp

-0.50

POSITIVE LOGITS

 common

1.95

 Common

1.84

common

1.79

Common

1.77

 COMMON

1.65

COMMON

1.48

 común

1.34

 comunes

1.28

 comum

1.28

 commons

1.21

Activations Density 0.141%

references to common concepts or items

No Comments

No Known Activations