INDEX

Explanations

references to specific geographic locations, particularly in China

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

Ops

-0.16

interop

-0.15

utra

-0.15

ÎµÏĤ

-0.14

rypton

-0.14

itty

-0.14

ÃĹ↵↵

-0.14

eso

-0.14

amide

-0.14

POSITIVE LOGITS

dong

0.23

zhou

0.21

urm

0.18

 Lumpur

0.15

ÅŁi

0.15

ilik

0.14

Ø¶ÙĬ

0.14

liner

0.14

±Ð¾ÑĤ

0.14

Activations Density 0.003%

references to specific geographic locations, particularly in China

No Comments

No Known Activations