INDEX

Explanations

statistics and numerical data related to performance metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

abad

-0.15

ukan

-0.14

ulia

-0.14

 Third

-0.13

strand

-0.13

adden

-0.13

ÙĨØ¯Ùĩ

-0.13

illes

-0.13

wargs

-0.13

iner

-0.13

POSITIVE LOGITS

 total

0.22

 altogether

0.19

total

0.18

 possible

0.18

possible

0.18

 those

0.17

 last

0.17

-total

0.17

 past

0.16

those

0.16

Activations Density 0.045%

statistics and numerical data related to performance metrics

No Comments

No Known Activations