Model Details
This model is a 1B llama3 model pretrained from scratch with torchtitan on fineweb-edu with AdamW optimizer. 100B tokens seen.
How to use
import torch
from transformers import pipeline
pipe = pipeline(
"text-generation",
model="kz919/llama3_1b_cautious_100B_token_8222025",
)
print(pipe("The key to life is"))
Downstream Eval
ARC, Hellaswag, Lambda_OpenAI, OpenbookQA, PIQA
lm_eval --model hf --model_args pretrained=kz919/llama3_1b_cautious_100B_token_8222025,dtype="bfloat16",add_bos_token=True --tasks lambada_openai,hellaswag,piqa,arc_easy,arc_challenge,openbookqa --device cuda:7 --batch_size 8
| Tasks |
Version |
Filter |
n-shot |
Metric |
|
Value |
|
Stderr |
| arc_challenge |
1 |
none |
0 |
acc |
↑ |
0.3123 |
± |
0.0135 |
|
|
none |
0 |
acc_norm |
↑ |
0.3413 |
± |
0.0139 |
| arc_easy |
1 |
none |
0 |
acc |
↑ |
0.6768 |
± |
0.0096 |
|
|
none |
0 |
acc_norm |
↑ |
0.5922 |
± |
0.0101 |
| hellaswag |
1 |
none |
0 |
acc |
↑ |
0.4007 |
± |
0.0049 |
|
|
none |
0 |
acc_norm |
↑ |
0.5092 |
± |
0.0050 |
| lambada_openai |
1 |
none |
0 |
acc |
↑ |
0.3901 |
± |
0.0068 |
|
|
none |
0 |
perplexity |
↓ |
21.6290 |
± |
0.7689 |
| openbookqa |
1 |
none |
0 |
acc |
↑ |
0.2660 |
± |
0.0198 |
|
|
none |
0 |
acc_norm |
↑ |
0.3680 |
± |
0.0216 |
| piqa |
1 |
none |
0 |
acc |
↑ |
0.7127 |
± |
0.0106 |
|
|
none |
0 |
acc_norm |
↑ |
0.7100 |
± |
0.0106 |
MMLU
| Groups |
Version |
Filter |
n-shot |
Metric |
|
Value |
|
Stderr |
| mmlu |
2 |
none |
|
acc |
↑ |
0.2515 |
± |
0.0037 |
| - humanities |
2 |
none |
|
acc |
↑ |
0.2451 |
± |
0.0063 |
| - other |
2 |
none |
|
acc |
↑ |
0.2716 |
± |
0.0080 |
| - social sciences |
2 |
none |
|
acc |
↑ |
0.2476 |
± |
0.0078 |
| - stem |
2 |
none |
|
acc |
↑ |
0.2452 |
± |
0.0076 |
