INDEX

Explanations

terms related to health and fitness categories

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

 Sections

-0.87

Sections

-0.84

 Phases

-0.82

 Regions

-0.81

Phases

-0.77

 Stages

-0.75

 Segments

-0.74

Stages

-0.73

 Seasons

-0.73

 Periods

-0.73

POSITIVE LOGITS

 autorytatywna

0.65

 sector

0.61

 category

0.57

 section

0.52

 chapter

0.50

 phase

0.49

 segment

0.47

 faction

0.44

 genre

0.43

 inning

0.43

Activations Density 1.573%

terms related to health and fitness categories

No Comments

No Known Activations