INDEX

Explanations

references to tracking concepts or mechanisms

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 Morley

-0.86

ngOnInit

-0.81

 Maier

-0.77

المناصب

-0.76

 Valeria

-0.72

 newValue

-0.72

 Dilemma

-0.71

 écl

-0.70

 Bolts

-0.69

alló

-0.69

POSITIVE LOGITS

 tracks

1.66

 Track

1.64

 Tracks

1.63

 TRACK

1.62

 track

1.61

Track

1.53

TRACK

1.49

track

1.48

tracks

1.46

Tracks

1.37

Activations Density 0.055%

references to tracking concepts or mechanisms

No Comments

No Known Activations