INDEX

Explanations

references to research studies or citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ufe

-0.16

 directions

-0.16

ãĥ³ãĥĩãĤ£

-0.15

andler

-0.14

mah

-0.14

eng

-0.14

ti

-0.14

 butt

-0.14

 quarter

-0.14

 bones

-0.14

POSITIVE LOGITS

 Coil

0.23

 Genel

0.20

 Coupon

0.18

inki

0.17

 Lilly

0.17

letal

0.17

 Govern

0.16

baugh

0.16

Dek

0.16

tec

0.16

Activations Density 0.001%

references to research studies or citations

No Comments

No Known Activations