INDEX

Explanations

references to programming concepts and technical processes

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

bum

-0.15

ãģ¾ãĤĮ

-0.14

.struts

-0.13

egas

-0.13

/Delete

-0.13

ito

-0.13

_strdup

-0.13

Accessibility

-0.13

 messaging

-0.13

POSITION

-0.12

POSITIVE LOGITS

FFE

0.18

ambi

0.17

 useRouter

0.15

akhir

0.14

yal

0.14

icer

0.14

bang

0.14

Techn

0.13

mie

0.13

use

0.13

Activations Density 0.134%

references to programming concepts and technical processes

No Comments

No Known Activations