INDEX

Explanations

specific terms related to identity and truth within various contexts

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

foreign

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

expandindo

-0.59

 Diſ

-0.53

 autorytatywna

-0.49

 فريبيس

-0.48

ArrowToggle

-0.46

 quæ

-0.44

 informée

-0.44

TypedDataSet

-0.43

原始内容存档于

-0.43

localctx

-0.42

POSITIVE LOGITS

VYMaps

0.44

 nakalista

0.41

Atoi

0.37

 معم

0.37

 bambú

0.35

Pranala

0.35

 dikel

0.35

клопе

0.34

 kasarigan

0.34

 otomatig

0.33

Activations Density 0.187%

specific terms related to identity and truth within various contexts

foreign

No Comments

No Known Activations