INDEX

Explanations

different cases and categories of letters in text

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

lowercase text

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

 propOrder

-0.62

 dieſer

-0.61

verwijspagina

-0.60

rungsseite

-0.60

 deſſen

-0.59

transQ

-0.59

 BoxFit

-0.58

 ſeinem

-0.58

[@BOS@]

-0.57

<unused8>

-0.57

POSITIVE LOGITS

Cheap

0.38

LowerCase

0.38

 intptr

0.37

lowercase

0.37

 lowercase

0.35

qrstuvwxyz

0.35

 depressed

0.32

不起

0.32

strtolower

0.32

dė

0.32

Activations Density 0.016%

different cases and categories of letters in text

lowercase text

No Comments

No Known Activations