INDEX

Explanations

phrases related to prognosis and medical outcomes

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

technical or scientific terms and metrics, particularly related to research, computing, and data analysis.

oai_token-act-pair · claude-3-5-haiku-20241022 Triggered by @neilrathi

This neuron detects phrases starting with "of" followed by nouns, particularly in technical or academic contexts.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

technical or scientific measurements, quantities, or percentages.

oai_token-act-pair · gpt-4o Triggered by @neilrathi

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

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

pgterms

-0.41

 rather

-0.39

 happy

-0.38

はじめまして

-0.37

 conf

-0.37

!*\

-0.36

°.

-0.36

randrange

-0.35

™.

-0.35

 remarqué

-0.35

POSITIVE LOGITS

 kağıdı

0.50

 tästä

0.49

 theſe

0.48

 titolata

0.47

makeText

0.47

 desenli

0.47

 kysy

0.47

Personendaten

0.46

 ſont

0.46

pecabe

0.45

Activations Density 0.620%

phrases related to prognosis and medical outcomes

technical or scientific terms and metrics, particularly related to research, computing, and data analysis.

This neuron detects phrases starting with "of" followed by nouns, particularly in technical or academic contexts.

technical or scientific measurements, quantities, or percentages.

No Comments

No Known Activations