INDEX

Explanations

words and concepts associated with legitimacy and authenticity in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

KommentareTeilen

-0.58

########.

-0.57

 aguda

-0.52

 balanceOf

-0.51

 GenerationType

-0.50

vezető

-0.48

 Emits

-0.47

lük

-0.46

RegistryLite

-0.46

 juntas

-0.44

POSITIVE LOGITS

化

1.81

ization

1.78

isation

1.62

izing

1.44

化的

1.43

ized

1.42

 hóa

1.40

IZATION

1.34

izations

1.34

ize

1.32

Activations Density 0.455%

words and concepts associated with legitimacy and authenticity in various contexts

No Comments

No Known Activations