INDEX

Explanations

references to historical events and geographic information

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

\Modules

-0.16

 chim

-0.15

 Tick

-0.15

 Chim

-0.15

zano

-0.15

.Tick

-0.14

lettes

-0.14

achts

-0.13

wf

-0.13

 roÄįnÃŃ

-0.13

POSITIVE LOGITS

dens

0.15

ãģ¡ãģ¯

0.14

ÑĥÑĢÐ°

0.14

Ø¬Ùĩ

0.14

ips

0.14

Ga

0.14

VOKE

0.14

UCK

0.14

Elo

0.14

kla

0.13

Activations Density 0.023%

references to historical events and geographic information

No Comments

No Known Activations