INDEX

Explanations

references to relationships and connection between people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

 immediately

-0.82

 inmediatamente

-0.82

 imediatamente

-0.81

immediately

-0.81

 instantly

-0.80

 inmediata

-0.79

Immediately

-0.79

 aussitôt

-0.75

 Immediately

-0.71

 Instantly

-0.70

POSITIVE LOGITS

 fully

0.79

 settled

0.73

 settle

0.71

 settles

0.71

 stabilized

0.70

 Fully

0.68

 stabilize

0.68

 stabili

0.68

Fully

0.66

 decide

0.63

Activations Density 0.687%

references to relationships and connection between people

No Comments

No Known Activations