INDEX

Explanations

elements related to specific organizations or individuals involved in medical or therapeutic contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

sidemargin

-0.63

 AssemblyCulture

-0.55

KommentareTeilen

-0.51

endpush

-0.47

theon

-0.46

arrings

-0.45

allible

-0.44

日閲覧

-0.44

AnchorTagHelper

-0.43

 fieldNum

-0.43

POSITIVE LOGITS

clerView

0.49

 pleaſure

0.48

 faſt

0.47

 ſche

0.46

 ſta

0.45

 itſelf

0.45

 ſever

0.45

 sumpay

0.43

 fernández

0.42

 yoox

0.42

Activations Density 1.406%

elements related to specific organizations or individuals involved in medical or therapeutic contexts

No Comments

No Known Activations