INDEX

Explanations

topics related to systematic analysis and evaluation in various fields

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

@nate

-0.15

reek

-0.15

osg

-0.14

æĻ

-0.14

 loose

-0.14

olla

-0.14

å¤ªéĥİ

-0.14

orus

-0.13

[__

-0.13

 Bilg

-0.13

POSITIVE LOGITS

ActionCreators

0.16

amura

0.15

ickle

0.15

yon

0.14

Ð½Ð¸Ñİ

0.14

oul

0.14

SND

0.13

ipple

0.13

Pam

0.13

its

0.13

Activations Density 0.086%

topics related to systematic analysis and evaluation in various fields

No Comments

No Known Activations