INDEX

Explanations

statistical data and numerical trends related to growth and comparison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

otine

-0.15

kaz

-0.15

olina

-0.15

Ø±Ø®

-0.14

 dipping

-0.13

owie

-0.13

erer

-0.13

ÏĥÏĦÎ®

-0.13

jug

-0.13

zl

-0.13

POSITIVE LOGITS

 triple

0.53

 quadr

0.52

 quint

0.46

tre

0.44

 nearly

0.44

 double

0.43

 triples

0.41

 doubled

0.41

 Triple

0.41

 doubles

0.41

Activations Density 0.097%

statistical data and numerical trends related to growth and comparison

No Comments

No Known Activations