INDEX

Explanations

verbs and phrases indicating knowledge or familiarity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 vuitton

-0.63

깥

-0.56

 betreft

-0.51

zeera

-0.50

 Photocase

-0.49

esgue

-0.49

ImageField

-0.48

 razie

-0.47

indelijk

-0.47

erialize

-0.47

POSITIVE LOGITS

 known

1.66

 Known

1.59

known

1.56

Known

1.55

 KNOWN

1.44

 γνω

1.34

KNOWN

1.32

 conhecido

1.32

 connu

1.32

 conocido

1.31

Activations Density 0.186%

verbs and phrases indicating knowledge or familiarity

No Comments

No Known Activations