INDEX

Explanations

punctuation marks and ellipses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

vim

-0.15

reib

-0.15

erb

-0.15

ALLERY

-0.14

lands

-0.14

æ

-0.13

vale

-0.13

 æĥ

-0.13

éĤ

-0.13

ades

-0.13

POSITIVE LOGITS

ponible

0.18

ean

0.17

owy

0.16

LOAT

0.15

OfWork

0.15

uards

0.15

ahlen

0.15

-fontawesome

0.14

ä¸Ī

0.14

URLException

0.14

Activations Density 0.006%

punctuation marks and ellipses

No Comments

No Known Activations