INDEX

Explanations

technical terminology related to medical conditions and patient care

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

Followed often by a preposition

np_acts-logits-general · gemini-2.0-flash

often, usually, typically

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

#+#

-0.44

Bra

-0.40

bra

-0.40

 Barre

-0.40

isNameExpr

-0.40

Bra

-0.39

belea

-0.39

ocyclic

-0.39

DockStyle

-0.39

äste

-0.38

POSITIVE LOGITS

 spesso

0.68

 usually

0.66

 vaak

0.65

 zwykle

0.63

 typically

0.63

 suele

0.61

 souvent

0.60

 Usually

0.59

Usually

0.59

 może

0.59

Activations Density 1.096%

technical terminology related to medical conditions and patient care

Followed often by a preposition

often, usually, typically

No Comments

No Known Activations