INDEX

Explanations

phrases related to historical landmarks and their features

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

errick

-0.15

unami

-0.15

 váºŃy

-0.15

mittel

-0.14

witch

-0.14

 Julien

-0.14

 Ð¿ÑĢÐ¸Ñĩ

-0.14

ifers

-0.13

pg

-0.13

PG

-0.13

POSITIVE LOGITS

elter

0.15

åĥį

0.14

atk

0.14

iphy

0.14

 Bott

0.14

,readonly

0.13

izon

0.13

achment

0.13

olis

0.13

Î²Î±

0.13

Activations Density 0.217%

phrases related to historical landmarks and their features

No Comments

No Known Activations