INDEX

Explanations

various instances of names and naming decisions throughout the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

opath

-0.15

-rated

-0.15

illos

-0.14

 Ratings

-0.14

elow

-0.14

-peer

-0.14

 axiom

-0.13

opathic

-0.13

è¯Ŀ

-0.13

ucher

-0.13

POSITIVE LOGITS

hy

0.22

 chosen

0.20

 shorten

0.18

urally

0.18

 descriptive

0.18

 Adopt

0.17

 adopted

0.17

 ÙĨØ³Ø¨Ø©

0.17

éŁ¿

0.17

pun

0.17

Activations Density 0.170%

various instances of names and naming decisions throughout the document

No Comments

No Known Activations