INDEX

Explanations

highly specific numerical or mathematical expressions and constructs

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

er

-0.19

ed

-0.18

zburg

-0.17

in

-0.17

-0.16

zelf

-0.15

Ø§

-0.15

anmar

-0.15

POSITIVE LOGITS

ndef

0.16

ereo

0.16

umer

0.15

nat

0.15

angu

0.14

uring

0.13

icont

0.13

 Robbins

0.13

Éµ

0.13

uby

0.13

Activations Density 0.201%

highly specific numerical or mathematical expressions and constructs

No Comments

No Known Activations

highly specific numerical or mathematical expressions and constructs

No Comments

No Known Activations