INDEX

Explanations

references to editors and editorial roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ness

-0.15

ayers

-0.15

ey

-0.15

eners

-0.14

_dll

-0.14

avy

-0.14

PARAM

-0.14

es

-0.14

nesc

-0.14

POSITIVE LOGITS

ials

0.43

ial

0.40

ially

0.35

-in

0.34

IAL

0.31

iale

0.30

ialized

0.28

ship

0.25

-at

0.24

iales

0.24

Activations Density 0.018%

references to editors and editorial roles

No Comments

No Known Activations