INDEX

Explanations

references to global concepts or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 défend

-0.97

 rencontr

-0.83

 étoit

-0.72

chism

-0.70

 RIAA

-0.70

 différente

-0.68

 doigts

-0.68

 presidencial

-0.68

 disputa

-0.67

 PyQt

-0.66

POSITIVE LOGITS

 global

1.19

 Global

1.19

GLOBAL

1.14

 GLOBAL

1.13

lobal

0.95

globals

0.94

globalData

0.94

Global

0.93

 globales

0.92

 חיצוניים

0.92

Activations Density 0.051%

references to global concepts or systems

No Comments

No Known Activations