INDEX

Explanations

numerical data and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 trio

-0.15

Î¼Î²

-0.14

 VÅ¡

-0.14

BOSE

-0.14

none

-0.14

 theirs

-0.14

 THREE

-0.13

ÐºÑĥÑĤ

-0.13

ycler

-0.13

``(

-0.13

POSITIVE LOGITS

Both

0.34

both

0.33

 both

0.33

 Both

0.32

 BOTH

0.31

 ambos

0.28

_both

0.27

 beide

0.26

_BOTH

0.22

 Ð¾Ð±Ð¾

0.22

Activations Density 0.125%

numerical data and comparisons

No Comments

No Known Activations