INDEX

Explanations

the mathematical notation for functions or variables labeled with 'l'

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

iama

-0.80

awake

-0.76

 attemp

-0.74

aed

-0.71

Hæ

-0.69

embed

-0.69

${{

-0.69

ruptedException

-0.67

 seismo

-0.67

 Narod

-0.67

POSITIVE LOGITS

1.33

1.15

getL

1.12

1.09

hl

1.03

ｌ

1.02

gl

0.99

isl

0.99

erl

0.95

Activations Density 0.233%

the mathematical notation for functions or variables labeled with 'l'

No Comments

No Known Activations