INDEX

Explanations

scientific terminology and notations

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

LaTeX math commands and `we`

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

-0.47

 дописавши

-0.42

 necesar

-0.40

 JpaRepository

-0.37

gonic

-0.37

pointments

-0.36

ardless

-0.36

 blowing

-0.35

 oorlog

-0.35

吹

-0.35

POSITIVE LOGITS

 pleaſure

0.58

 ſeveral

0.56

 houſe

0.54

 myſelf

0.54

 ſte

0.53

 faſt

0.52

 Houſe

0.52

 ſmall

0.52

WithIOException

0.50

 ſub

0.49

Activations Density 0.001%

scientific terminology and notations

LaTeX math commands and `we`

No Comments

No Known Activations