INDEX

Explanations

phrases that indicate something is valuable or significant

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 وتسجيلات

-0.66

ViewFeatures

-0.60

tonsoft

-0.59

]--;

-0.58

 المعيارى

-0.58

BindView

-0.57

 gynhyrchwyd

-0.57

almaz

-0.57

VIER

-0.56

 Fische

-0.56

POSITIVE LOGITS

 worth

4.28

 Worth

3.60

 WORTH

3.52

Worth

3.43

worth

3.21

WORTH

2.90

 worthwhile

2.14

 worthy

1.99

值得

1.74

 Worthy

1.73

Activations Density 0.053%

phrases that indicate something is valuable or significant

No Comments

No Known Activations