INDEX

Explanations

locations and geographic identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĥn

-0.18

imon

-0.18

uras

-0.16

oise

-0.16

Ã§os

-0.16

pas

-0.16

ernet

-0.15

å«Į

-0.15

urent

-0.15

-0.14

POSITIVE LOGITS

iris

0.20

mium

0.18

(OS

0.18

åıİ

0.17

.environ

0.17

oba

0.17

.path

0.16

mos

0.16

bourne

0.16

abei

0.16

Activations Density 0.024%

locations and geographic identifiers

No Comments

No Known Activations