INDEX

Explanations

terms related to exceptional quality or performance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

immers

-0.15

agar

-0.15

isphere

-0.14

zik

-0.14

ode

-0.14

ambi

-0.14

 then

-0.14

anel

-0.14

 Religion

-0.13

POSITIVE LOGITS

-quality

0.18

ively

0.17

iyi

0.15

 sayÄ±da

0.15

/conf

0.14

BackColor

0.14

alley

0.14

itom

0.14

.runners

0.14

tÃ¼

0.14

Activations Density 0.009%

terms related to exceptional quality or performance

No Comments

No Known Activations