INDEX

Explanations

actions related to cutting and preparing food

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

lint

-0.16

 Socket

-0.15

opot

-0.15

socket

-0.15

veis

-0.15

Ã¡ria

-0.14

unden

-0.14

ElementException

-0.14

tica

-0.14

æĭĶ

-0.14

POSITIVE LOGITS

jul

0.27

 thickness

0.26

 thin

0.26

 thick

0.25

 length

0.24

 uniform

0.22

 coins

0.22

 Thickness

0.22

 rounds

0.22

 strips

0.22

Activations Density 0.022%

actions related to cutting and preparing food

No Comments

No Known Activations