INDEX

Explanations

references to application structure and interface elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

odial

-0.14

 Citizen

-0.14

ioned

-0.14

Dyn

-0.14

oder

-0.14

 Rank

-0.13

Ø§ÙĤ

-0.13

chaft

-0.13

ise

-0.13

odel

-0.13

POSITIVE LOGITS

-wide

0.20

 Equip

0.16

Equip

0.15

ubb

0.15

wide

0.14

ÎµÎ¹Î±

0.14

ldr

0.14

entiful

0.14

context

0.14

uide

0.14

Activations Density 0.105%

references to application structure and interface elements

No Comments

No Known Activations