INDEX

Explanations

patterns and structures related to programming and data formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

nave

-0.15

ä¸ģ

-0.14

Ŀ

-0.14

ess

-0.14

 åĪ

-0.13

 trap

-0.13

trap

-0.13

_streams

-0.13

å¸Ń

-0.13

ington

-0.13

POSITIVE LOGITS

icontrol

0.15

hlen

0.15

ahlen

0.15

ISCO

0.15

velop

0.15

 arac

0.15

ARAM

0.14

bout

0.14

avra

0.14

ctest

0.14

Activations Density 0.031%

patterns and structures related to programming and data formatting

No Comments

No Known Activations