INDEX

Explanations

programming constructs and components

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

code and citations

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

ftagPool

-0.43

MarshalTo

-0.42

reactivex

-0.40

RuleContext

-0.40

findpost

-0.39

 laude

-0.38

labelledby

-0.37

ParallelGroup

-0.36

cloudflare

-0.35

 חיצוניים

-0.35

POSITIVE LOGITS

 betweenstory

0.66

rungsseite

0.65

 RIPRODUZIONE

0.57

 يتيمه

0.56

 plufieurs

0.54

 Италијани

0.54

 jsPsych

0.54

BibitemShut

0.53

 iVar

0.52

 ſch

0.52

Activations Density 0.012%

programming constructs and components

code and citations

No Comments

No Known Activations