INDEX

Explanations

content related to events and dates

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

 Morav

-0.16

ysqli

-0.15

andel

-0.15

ippi

-0.14

terior

-0.14

hyth

-0.14

sap

-0.14

owski

-0.14

 Illustrated

-0.14

á»¯u

-0.14

POSITIVE LOGITS

ote

0.17

Ä±ÄŁÄ±

0.16

aset

0.16

osi

0.15

inder

0.15

Pra

0.15

chu

0.14

 Torch

0.14

erval

0.14

/vnd

0.14

Activations Density 0.075%

content related to events and dates

No Comments

No Known Activations