INDEX

Explanations

phrases related to medical records and health issues

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

things overlooked or incorrectly identified

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

refs

-0.43

sodik

-0.41

 שוליים

-0.40

blico

-0.39

ref

-0.38

wave

-0.38

スタンド

-0.38

 Venise

-0.38

back

-0.38

 reprend

-0.38

POSITIVE LOGITS

AddTagHelper

0.50

 Wikimedijinoj

0.48

 vergessen

0.46

 saites

0.45

 wcześ

0.44

Vidite

0.43

 overlooked

0.40

 forgotten

0.40

inWeight

0.39

 olvido

0.39

Activations Density 0.164%

phrases related to medical records and health issues

things overlooked or incorrectly identified

No Comments

No Known Activations