INDEX

Explanations

references to educational topics or practices

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

à¸Ļà¸Ķ

-0.12

_TestCase

-0.12

reau

-0.12

anford

-0.12

NetMessage

-0.11

azen

-0.11

.Cursors

-0.11

ibel

-0.11

getDoctrine

-0.11

Ð»ÐµÐ¼

-0.11

POSITIVE LOGITS

 multi

1.09

 Multi

1.05

 mult

0.98

 Mult

0.97

multi

0.96

 MULTI

0.95

Multi

0.94

 MULT

0.93

-multi

0.92

mult

0.92

Activations Density 0.311%

references to educational topics or practices

No Comments

No Known Activations