INDEX

Explanations

phrases that suggest improvement or positive transformations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

estris

-0.53

 sièges

-0.52

UTERS

-0.50

 uintptr

-0.49

'{@

-0.48

 '\\;'

-0.48

iD

-0.47

źć

-0.47

 vanguardia

-0.47

 بيها

-0.46

POSITIVE LOGITS

 propOrder

0.70

Bakgrunnsstoff

0.65

enderror

0.60

 AppBar

0.53

contentLoaded

0.53

 Majefty

0.52

限り

0.52

 alfab

0.52

engkapi

0.51

parsedMessage

0.49

Activations Density 0.094%

phrases that suggest improvement or positive transformations

No Comments

No Known Activations