INDEX

Explanations

terms and phrases related to classification and categorization processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 aéri

-0.84

bagno

-0.83

 Tacitus

-0.81

argout

-0.80

 bParam

-0.79

 Notting

-0.78

writeHead

-0.78

uinal

-0.77

Linq

-0.77

jPanel

-0.77

POSITIVE LOGITS

 instruments

0.96

 Classi

0.89

Instruments

0.86

 Instruments

0.83

 instrument

0.82

 Instrument

0.80

Classify

0.79

 instrumental

0.78

 Herald

0.78

 classification

0.75

Activations Density 0.093%

terms and phrases related to classification and categorization processes

No Comments

No Known Activations