INDEX

Explanations

terms related to health and medical conditions, particularly focusing on pharmaceuticals and treatments

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 فريبيس

-0.63

AndEndTag

-0.60

 sekal

-0.60

FieldBuilder

-0.60

 Италијани

-0.58

 հղումներ

-0.58

 arşivlendi

-0.57

сылкі

-0.56

IntoConstraints

-0.56

 للمعارف

-0.55

POSITIVE LOGITS

heads

0.40

mapView

0.36

Ifc

0.35

 ſch

0.34

Enabling

0.33

orie

0.31

posts

0.31

 ואת

0.31

Draw

0.31

Walk

0.31

Activations Density 0.610%

terms related to health and medical conditions, particularly focusing on pharmaceuticals and treatments

No Comments

No Known Activations