INDEX

Explanations

best way

np_token-act-pair-logits · gemini-2.0-flash

phrases expressing superlative degree

oai_token-act-pair · gemini-2.0-flash

best, easiest, important

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_16/width_16k/average_l0_10

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pinulongan

-0.68

principalColumn

-0.61

ۜ

-0.59

 ControllerBase

-0.56

XmlAccessType

-0.55

 kasarigan

-0.55

 transfieras

-0.54

BeginContext

-0.53

ManagedBean

-0.53

tagHelperRunner

-0.52

POSITIVE LOGITS

AddField

0.57

AutoresizingMask

0.56

󠁴

0.54

PageIndex

0.51

juvant

0.49

conductor

0.49

addChild

0.49

 القدم

0.48

Referencies

0.48

loginForm

0.48

Activations Density 0.797%

best way

phrases expressing superlative degree

best, easiest, important

No Comments

No Known Activations