INDEX

Explanations

descriptions related to the characteristics and nature of objects or phenomena

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

↵↵

-0.54

↵

-0.54

di

-0.49

-0.47

“

-0.45

räck

-0.44

 These

-0.44

ivable

-0.43

Precise

-0.43

POSITIVE LOGITS

 GenerationType

1.12

 CreateTagHelper

0.95

 فريبيس

0.92

 disponibilités

0.87

 AspNetCore

0.87

UnsafeEnabled

0.84

IntoConstraints

0.84

printStackTrace

0.83

󠁴

0.82

 صوتيه

0.81

Activations Density 0.637%

descriptions related to the characteristics and nature of objects or phenomena

No Comments

No Known Activations