INDEX

Explanations

information related to medical professionals and healthcare contexts

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

earned his undergraduate

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



-0.66

GTCX

-0.61

 estekak

-0.60

ⓧ

-0.60

verwijspagina

-0.57

новниш

-0.56

 незавершена

-0.56

 Numerade

-0.55

 snippetHide

-0.54

 GenerationType

-0.52

POSITIVE LOGITS

 this

0.36

 pontu

0.34

flashdata

0.33

 Erster

0.33

ఔ

0.32

 Bowman

0.32

 uved

0.31

señor

0.31

 erster

0.31

 longer

0.31

Activations Density 0.084%

information related to medical professionals and healthcare contexts

earned his undergraduate

No Comments

No Known Activations