INDEX

Explanations

references to trips or travel-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

AndEndTag

-0.69

 Fixture

-0.61

 appointment

-0.61

 ultras

-0.59

himovic

-0.59

 Intersection

-0.56

 propOrder

-0.56

 emoji

-0.56

 porcentual

-0.56

 memo

-0.56

POSITIVE LOGITS

tho

0.45

 Simplemente

0.43

 transfieras

0.41

 clásica

0.41

 ddelweddau

0.41

 klassische

0.41

 gambe

0.40

 perfeitamente

0.39

 egentligen

0.39

 pengantin

0.39

Activations Density 0.159%

references to trips or travel-related terms

No Comments

No Known Activations