INDEX

Explanations

terms related to medical conditions and treatments

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Zweiten

-0.44

OGND

-0.41

 positivas

-0.39

 potrivit

-0.39

ására

-0.35

ssaint

-0.35

іод

-0.34

 אות

-0.33

 orgull

-0.33

正

-0.33

POSITIVE LOGITS



0.59

 ब्रेकडाउन

0.57

Халык

0.57

ISupport

0.51

 насељу

0.48

tetlen

0.47

 surla

0.47

 navigation

0.45

 Taktlose

0.45

HandlerContext

0.45

Activations Density 1.029%

terms related to medical conditions and treatments

No Comments

No Known Activations