INDEX

Explanations

properties and attributes in a structured or code format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

cade

-0.20

ama

-0.19

RefCount

-0.15

SY

-0.15

viz

-0.15

eda

-0.14

RIEND

-0.14

agina

-0.14

adin

-0.14

 çľ

-0.14

POSITIVE LOGITS

elps

0.15

 Donovan

0.14

/messages

0.14

Peg

0.14

sst

0.14

Ãºc

0.14

itm

0.14

ãģĭãĤı

0.14

Activations Density 0.028%

properties and attributes in a structured or code format

No Comments

No Known Activations