INDEX

Explanations

references to specific organizations or events related to computer science

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

erialize

-0.18

eu

-0.17

oles

-0.17

orsch

-0.17

TY

-0.16

elf

-0.16

ecurity

-0.15

oct

-0.15

oft

-0.15

POSITIVE LOGITS

IRO

0.24

/cs

0.17

CS

0.16

atica

0.16

Cs

0.15

utom

0.15

 Lewis

0.15

irt

0.15

 Ekon

0.15

rollo

0.14

Activations Density 0.015%

references to specific organizations or events related to computer science

No Comments

No Known Activations