INDEX

Explanations

negative attributes and concepts related to health and well-being

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

Française

-0.62

 telle

-0.59

 fofo

-0.58

openConnection

-0.57

__':

-0.57

 Albans

-0.56

 stabilize

-0.55

 altos

-0.55

 Specialization

-0.55

ята

-0.54

POSITIVE LOGITS

 själ

0.74

 ویکی‌پدیا

0.73

 defaultstate

0.72

tvguidetime

0.69

طبيق

0.69

+#+#

0.66

roskop

0.65

 actionTypes

0.65

TIN

0.62

ousand

0.61

Activations Density 0.878%

negative attributes and concepts related to health and well-being

No Comments

No Known Activations