INDEX

Explanations

key terms and variables related to programming and coding contexts

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

mathematical or archaic notations

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

AddHtmlAttribute

-0.65

uxxxx

-0.60

OrNil

-0.59

WillAppear

-0.57

 مرئيه

-0.56

lorette

-0.56

Personendaten

-0.56

ckså

-0.55

下载附件

-0.54

InstrumentedTest

-0.54

POSITIVE LOGITS

 tranſ

0.49

 inſ

0.48

 poffible

0.48

 Inſ

0.47

 ſur

0.46

 ſever

0.45

RTHOOK

0.44

 Reſ

0.43

 ſal

0.43

 ſel

0.41

Activations Density 0.060%

key terms and variables related to programming and coding contexts

mathematical or archaic notations

No Comments

No Known Activations