INDEX

Explanations

terms associated with diabetes and related medical conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Houſe

-0.61

 Üniversitesi

-0.61

 uſe

-0.55

Mereka

-0.51

ltä

-0.50

 houſe

-0.50

Źródło

-0.50

 Japón

-0.49

ſo

-0.49

ึ้น

-0.48

POSITIVE LOGITS

bolic

0.95

neurial

0.91

ratic

0.89

riotic

0.87

ronic

0.86

betic

0.86

ynthetic

0.86

olic

0.85

athic

0.84

hagic

0.83

Activations Density 0.192%

terms associated with diabetes and related medical conditions

No Comments

No Known Activations