INDEX

Explanations

phrases related to limitations and entries in promotions or giveaways

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

lando

-0.17

iel

-0.16

lland

-0.15

slaught

-0.15

EDIA

-0.15

Ð½Ð¸Ð¼Ð°

-0.14

esor

-0.14

æī£

-0.14

 springfox

-0.14

ikit

-0.14

POSITIVE LOGITS

 multiple

0.17

 Multiple

0.17

 basil

0.16

 maximum

0.15

 Maximum

0.15

_MAX

0.15

Isl

0.15

 Ð±Ð¾Ð¹

0.15

max

0.15

 Twice

0.15

Activations Density 0.042%

phrases related to limitations and entries in promotions or giveaways

No Comments

No Known Activations