INDEX

Explanations

phrases that express similarity or comparison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

bilità

-0.32

 verband

-0.32

قال

-0.30

Bakgrunnsstoff

-0.29

removeAttribute

-0.28

ISODE

-0.27

 dignité

-0.27

❋

-0.27

 frumos

-0.27

 schop

-0.27

POSITIVE LOGITS

 pure

1.04

AndEndTag

0.92

pure

0.88

 overkill

0.86

 PURE

0.85

 pura

0.83

 deja

0.82

 manna

0.80

 heaven

0.79

ConstraintMaker

0.77

Activations Density 0.446%

phrases that express similarity or comparison

No Comments

No Known Activations