INDEX

Explanations

phrases indicating temporal relationships or conditions involving independence and prior knowledge

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 Jefus

-0.96

 Anſ

-0.95

 greateſt

-0.90

 Reſ

-0.89

 Houſe

-0.88

 houſe

-0.88

 Conſ

-0.88

 Chriſt

-0.87

 Chriftian

-0.86

ſelf

-0.85

POSITIVE LOGITS

 independently

0.54

 CreateTagHelper

0.52

 already

0.51

ご了承ください

0.51

 unrelated

0.49

Already

0.47

 schon

0.47

 without

0.46

 separate

0.45

 than

0.45

Activations Density 0.411%

phrases indicating temporal relationships or conditions involving independence and prior knowledge

No Comments

No Known Activations