INDEX

Explanations

strong adjectives and descriptive phrases related to product quality and features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 '\\;'

-1.03

 <=",

-0.83

AsUp

-0.83

UnsafeEnabled

-0.82

 ExecuteAsync

-0.81

بوابة

-0.80

ValueStyle

-0.80

 Chwiliwch

-0.80

DataAnnotations

-0.75

 незавершена

-0.73

POSITIVE LOGITS

 manufactured

0.61

0.58

menea

0.53

manufactured

0.52

Features

0.50

 packs

0.49

dallah

0.48

 pack

0.48

 products

0.47

Top

0.47

Activations Density 0.223%

strong adjectives and descriptive phrases related to product quality and features

No Comments

No Known Activations