INDEX

Explanations

technical and statistical terminology related to research studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 feminina

-0.33

 sintética

-0.32

 geprü

-0.32

 gjø

-0.32

 tatlı

-0.31

internalType

-0.31

 Verbesserung

-0.31

 asiática

-0.30

 zorgen

-0.30

rimir

-0.30

POSITIVE LOGITS

 regression

1.07

 Regression

0.94

regression

0.85

 logistic

0.83

 statistical

0.81

 regressions

0.81

Regression

0.81

 linear

0.80

 AssemblyCulture

0.79

 Bayesian

0.78

Activations Density 1.954%

technical and statistical terminology related to research studies

No Comments

No Known Activations