INDEX

Explanations

numerical values and measurements, particularly related to statistics or data points

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ening

-0.20

ened

-0.18

scale

-0.17

eding

-0.17

ponent

-0.16

viders

-0.16

rosse

-0.15

wis

-0.15

owing

-0.15

ting

-0.15

POSITIVE LOGITS

ational

0.18

lew

0.17

ably

0.17

igel

0.17

ulumi

0.16

ÄŁan

0.16

IBUTE

0.16

lay

0.15

ful

0.15

OrUpdate

0.14

Activations Density 0.099%

numerical values and measurements, particularly related to statistics or data points

No Comments

No Known Activations