INDEX

Explanations

terms related to optimization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

rp

-0.18

led

-0.15

getc

-0.15

hound

-0.15

æŃ

-0.15

 Lair

-0.15

hurst

-0.14

leton

-0.14

hand

-0.14

iente

-0.14

POSITIVE LOGITS

imal

0.34

imum

0.32

ima

0.24

IMAL

0.23

Ð¸Ð¼Ð°Ð»ÑĮ

0.21

imize

0.21

imized

0.20

icks

0.20

ics

0.20

ician

0.20

Activations Density 0.011%

terms related to optimization

No Comments

No Known Activations