INDEX

Explanations

references to organizational acronyms and their activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 Prospect

-0.15

isbury

-0.15

wards

-0.15

plotlib

-0.15

urch

-0.15

ãģ¾ãģ¾

-0.15

Ã¼st

-0.14

roof

-0.14

wet

-0.14

Wet

-0.14

POSITIVE LOGITS

eland

0.26

_launcher

0.26

ICI

0.22

ICLES

0.21

arus

0.20

ymi

0.20

ANN

0.19

ICLE

0.18

icles

0.18

loud

0.17

Activations Density 0.008%

references to organizational acronyms and their activities

No Comments

No Known Activations

references to organizational acronyms and their activities

No Comments

No Known Activations