INDEX

Explanations

terms related to associations and patterns in biomedical contexts

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

associated structures or concepts

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

DotNetBar

-0.39

ChildScrollView

-0.35

 Exteriores

-0.35

hObject

-0.33

 discriminación

-0.33

opsida

-0.32

 timpul

-0.32

instancetype

-0.32

TestBed

-0.32

esistenza

-0.31

POSITIVE LOGITS

 Wikimedijinoj

0.51

gameserver

0.50

 Medium

0.49

 ExecuteAsync

0.48

IndentedString

0.47

transQ

0.47

 SafeMath

0.47

ByExample

0.46

 ashore

0.45

Medium

0.44

Activations Density 0.227%

terms related to associations and patterns in biomedical contexts

associated structures or concepts

No Comments

No Known Activations