INDEX

Explanations

numbers and references to mathematical or programming concepts

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

eland

-0.17

.wp

-0.14

zcze

-0.14

DOT

-0.14

 ÄĳoÃłn

-0.14

èŀ

-0.14

ileges

-0.14

oon

-0.13

itesse

-0.13

éĻ½

-0.13

POSITIVE LOGITS

osto

0.16

idental

0.15

veau

0.15

vens

0.14

 Carn

0.14

yan

0.14

abin

0.14

eneric

0.14

umm

0.13

 Flip

0.13

Activations Density 0.001%

numbers and references to mathematical or programming concepts

No Comments

No Known Activations