INDEX

Explanations

content related to historical figures and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

vsp

-0.20

 Ð´Ð¸Ð½Ð°

-0.18

enheim

-0.17

ortho

-0.16

abi

-0.15

UMB

-0.15

ohana

-0.15

lus

-0.14

ulty

-0.14

 ÑģÐµÐ»

-0.14

POSITIVE LOGITS

 vice

0.26

vice

0.23

 proton

0.23

 procur

0.23

 Podesta

0.23

 Vice

0.21

 chamber

0.20

 Chamber

0.19

vir

0.18

 govern

0.18

Activations Density 0.042%

content related to historical figures and events

No Comments

No Known Activations