INDEX

Explanations

terms indicating relative comparison or association

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/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.34.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

 AssemblyCulture

-0.66

 pinulongan

-0.65

 <<<<<<<<<<<<<<

-0.58

 Meksiku

-0.52

rungsseite

-0.50

tagHelperRunner

-0.50

 resourceCulture

-0.48

 Biôgrafia

-0.47

 kasarigan

-0.47

tvguidetime

-0.46

POSITIVE LOGITS

 Relative

0.66

 relative

0.65

Relative

0.59

relative

0.58

rel

0.53

 humidity

0.51

 speaking

0.49

 relativity

0.48

 relati

0.48

Rel

0.48

Activations Density 0.145%

terms indicating relative comparison or association

No Comments

No Known Activations

terms indicating relative comparison or association

No Comments

No Known Activations