INDEX

Explanations

references to books and numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

LiveData

-0.15

_refl

-0.15

rift

-0.15

ODB

-0.14

acam

-0.14

lingen

-0.14

orget

-0.14

 componentDid

-0.14

onz

-0.13

akk

-0.13

POSITIVE LOGITS

Ans

0.41

 ansible

0.40

ansible

0.40

Ans

0.36

 playbook

0.36

_ans

0.31

(ans

0.31

ans

0.29

 Play

0.28

ANS

0.28

Activations Density 0.004%

references to books and numerical data

No Comments

No Known Activations