INDEX

Explanations

phrases indicating suggestions or recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 McNally

-0.51

<strong>

-0.45

 nieruchomości

-0.44

 istrinya

-0.43

 mariposas

-0.43

 íntegra

-0.42

 kebenaran

-0.42

 travaillons

-0.42

Â

-0.41

 vloer

-0.41

POSITIVE LOGITS

 suggestion

1.35

 suggest

1.34

Suggest

1.30

 Suggestion

1.27

suggest

1.23

 Suggest

1.22

suggestion

1.21

Suggestion

1.20

 SUGGEST

1.19

 Suggestions

1.16

Activations Density 0.024%

phrases indicating suggestions or recommendations

No Comments

No Known Activations