INDEX

Explanations

numerical and statistical terms related to research and clinical studies

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits



-0.73

SBATCH

-0.69

 Infórmanos

-0.60

DockStyle

-0.59

 ujednoznacz

-0.59

 Stewardship

-0.56

lossians

-0.56

ropractic

-0.55

astify

-0.55

IsPostBack

-0.54

POSITIVE LOGITS

السكان

0.36

orianCalendar

0.34

 siff

0.33

 semblable

0.31

CompilerServices

0.31

vot

0.30

 universale

0.28

 wybor

0.28

 voting

0.27

cæ

0.27

Activations Density 1.137%

numerical and statistical terms related to research and clinical studies

No Comments

No Known Activations