INDEX

Explanations

references to books and reading materials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

uzz

-0.19

ors

-0.18

MENT

-0.17

brook

-0.17

ORS

-0.16

ugs

-0.16

zyst

-0.16

-0.15

igm

-0.15

POSITIVE LOGITS

keeping

0.26

ç±į

0.26

ends

0.26

shelf

0.26

worm

0.22

eller

0.20

ellers

0.20

stores

0.20

ended

0.20

shop

0.19

Activations Density 0.064%

references to books and reading materials

No Comments

No Known Activations