INDEX

Explanations

structures related to tabular data or mathematical notations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

isoft

-0.20

baugh

-0.17

rve

-0.16

esson

-0.16

iro

-0.15

etCode

-0.15

Ä©nh

-0.15

trl

-0.15

abeth

-0.15

ensen

-0.15

POSITIVE LOGITS

Iv

0.16

ModelProperty

0.16

idd

0.15

kea

0.15

\Abstract

0.14

ault

0.14

ISTA

0.14

Ð°Ð»Ñĭ

0.14

nex

0.14

à¹Ħà¸¥

0.14

Activations Density 0.005%

structures related to tabular data or mathematical notations

No Comments

No Known Activations