INDEX

Explanations

negations, things that are being avoided, details and facts that aren't needed or not being repeated

oai_token-act-pair · gemini-2.0-flash

Avoiding repetition/providing details

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_8/width_16k/average_l0_52

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+#+#

-0.84

DataAnnotations

-0.80

 '\\;'

-0.74

 propOrder

-0.73

Personendaten

-0.71

quelize

-0.65

OrNil

-0.64

Zeneca

-0.63

SpringBootTest

-0.63

AutoScaleMode

-0.61

POSITIVE LOGITS

engine

0.46

ANAL

0.46

詳

0.43

 destacados

0.42

 anal

0.41

 selesai

0.41

AsyncResult

0.41

cir

0.41

entina

0.40

 Golden

0.40

Activations Density 1.718%

negations, things that are being avoided, details and facts that aren't needed or not being repeated

Avoiding repetition/providing details

No Comments

No Known Activations