INDEX

Explanations

medical test results and healthcare references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

~$

-0.82

 poichè

-0.79

Whilst

-0.78

 houſe

-0.75

 ſch

-0.75

 ſever

-0.74

 Whilst

-0.72

 Monfieur

-0.71

 IIRC

-0.71

ſelves

-0.71

POSITIVE LOGITS

 antemano

0.41

NewLabel

0.40

enzial

0.39

 pály

0.37

Wich

0.37

 megy

0.37

돼

0.37

 masukan

0.36

 memas

0.35

 yanı

0.35

Activations Density 0.175%

medical test results and healthcare references

No Comments

No Known Activations