INDEX

Explanations

mentions of PBS and its affiliated entities or networks

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

ertz

-0.18

 Celt

-0.15

atri

-0.15

iland

-0.14

ixel

-0.14

oser

-0.14

acom

-0.14

Mev

-0.13

 Shut

-0.13

filtr

-0.13

POSITIVE LOGITS

 nationwide

0.18

endforeach

0.17

endi

0.17

tainment

0.16

éĶ

0.16

-equiv

0.15

 forall

0.15

 everywhere

0.15

kees

0.15

.od

0.14

Activations Density 0.600%

mentions of PBS and its affiliated entities or networks

No Comments

No Known Activations