INDEX

Explanations

terms related to healthcare interventions and evaluations

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

preceding adverbs of frequency

np_acts-logits-general · gemini-2.0-flash

usually, often, typically, oftentimes

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

 AssemblyCulture

-0.60

 cherchés

-0.58

 متعلقه

-0.56

このまま

-0.55

szön

-0.50

 তারিখ

-0.49

AddAttribute

-0.48

haupt

-0.48

webElementXpaths

-0.48

 calendriers

-0.48

POSITIVE LOGITS

 usually

3.81

 often

3.50

usually

3.38

 typically

3.29

 Usually

3.26

Usually

3.24

 Often

3.09

often

3.08

Often

3.01

 oftentimes

2.91

Activations Density 1.830%

terms related to healthcare interventions and evaluations

preceding adverbs of frequency

usually, often, typically, oftentimes

No Comments

No Known Activations