INDEX

Explanations

verbs and action-related phrases

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

 Raim

-0.73

 PopupWindow

-0.71

pnea

-0.71

HostException

-0.66

TintMode

-0.66

TextBoxColumn

-0.64

XtraGrid

-0.64

طيع

-0.62

 cikin

-0.62

 Arta

-0.60

POSITIVE LOGITS

 yourselves

0.97

 yourself

0.94

 themselves

0.86

yourself

0.85

::{

0.84

 ourselves

0.83

 himself

0.74

 oneself

0.72

Yourself

0.71

 lägga

0.69

Activations Density 0.200%

verbs and action-related phrases

No Comments

No Known Activations