INDEX

Explanations

terms related to health improvement and enhancement

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

"improve" or "improving"

np_acts-logits-general · gemini-2.0-flash

making things better

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 frumo

-0.45

⟬

-0.44

 Wicidata

-0.43

setColumns

-0.43

pesas

-0.42

 gehouden

-0.42

 nonzero

-0.42

高く

-0.41

 geweest

-0.41

 apellidos

-0.41

POSITIVE LOGITS

 quality

0.98

 efficiencies

0.96

 efficiency

0.94

 readability

0.86

quality

0.84

:✨

0.84

efficiency

0.84

 Efficiency

0.83

 upon

0.80

 odds

0.79

Activations Density 0.134%

terms related to health improvement and enhancement

"improve" or "improving"

making things better

No Comments

No Known Activations