INDEX

Explanations

instances of questions being asked

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

imony

-0.17

Ãºng

-0.15

á»īnh

-0.15

cles

-0.15

.nih

-0.15

arend

-0.14

iou

-0.14

 Ð°Ð¼

-0.14

owa

-0.14

iffin

-0.14

POSITIVE LOGITS

 Gors

0.15

Su

0.15

dbus

0.14

sav

0.14

bo

0.14

 chÃŃ

0.14

Cos

0.14

bc

0.14

Su

0.14

BC

0.13

Activations Density 0.182%

instances of questions being asked

No Comments

No Known Activations