INDEX

Explanations

terms related to clipboards and quantitative measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 surveyed

-0.41

 ipsum

-0.40

 exhaust

-0.38

 legend

-0.36

 Jamie

-0.35

 Honor

-0.35

 Legend

-0.34

 LEGEND

-0.34

 Glück

-0.34

Попис

-0.34

POSITIVE LOGITS

Cli

1.12

cli

1.06

Cli

1.04

cli

1.03

 quanti

1.00

 clip

0.97

 Quanti

0.96

Quanti

0.94

 clipping

0.91

 quantification

0.90

Activations Density 0.237%

terms related to clipboards and quantitative measurements

No Comments

No Known Activations