INDEX

Explanations

mathematical terms and expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

'Re

-0.15

 Sesso

-0.14

 GenÃ§

-0.13

Ð°Ð´ÐºÐ¸

-0.13

'S

-0.13

ãĢģãĢģ

-0.13

BoxLayout

-0.13

ãĥ¼ãĤ¸

-0.12

 Bilim

-0.12

POSITIVE LOGITS

 initials

0.16

 acronym

0.15

reatest

0.15

abbreviation

0.15

eren

0.15

.GetEnumerator

0.14

alten

0.14

enor

0.14

hani

0.13

 abbreviation

0.13

Activations Density 0.707%

mathematical terms and expressions

No Comments

No Known Activations