| .. |
|
base.yaml
|
71befc19e1
[Loss] Use flash_attn.losses.cross_entropy.CrossEntropyLoss
|
1 年之前 |
|
gpt3-2.7B-flash-8k.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3-2.7B-flash-hdim128-rotary-8k.yaml
|
c2407dec96
Fix typo in config: train.gpu -> train.gpu_mem
|
2 年之前 |
|
gpt3-2.7B-flash-hdim128-rotary.yaml
|
c2407dec96
Fix typo in config: train.gpu -> train.gpu_mem
|
2 年之前 |
|
gpt3-2.7B-flash-hdim128.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3-2.7B-flash-rotary-8k.yaml
|
c2407dec96
Fix typo in config: train.gpu -> train.gpu_mem
|
2 年之前 |
|
gpt3-2.7B-flash-rotary.yaml
|
c2407dec96
Fix typo in config: train.gpu -> train.gpu_mem
|
2 年之前 |
|
gpt3-2.7B-flash.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3-2.7B-hf-hdim128.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3-2.7B-hf.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3l-flash-8k.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3l-flash-rotary-30B.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3l-flash-rotary-8k.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3l-flash-rotary.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3l-flash.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3l-hf.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3m-flash-8k.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3m-flash-rotary-30B.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3m-flash-rotary-8k.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3m-flash-rotary.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3m-flash.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3m-hf.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3s-flash-8k.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3s-flash-rotary-30B.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3s-flash-rotary-8k.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3s-flash-rotary.yaml
|
0bf5e50038
Release training code
|
2 年之前 |
|
gpt3s-flash.yaml
|
88173a1aaf
[FusedDense] Support relu, rename FusedDenseGeluDense -> FusedMLP
|
1 年之前 |
|
gpt3s-hf.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3xl-flash-8k.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3xl-flash-rotary-60B.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3xl-flash-rotary-8k.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3xl-flash-rotary.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3xl-flash.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
|
gpt3xl-hf.yaml
|
4a6eaa9f27
Update configs, add results
|
2 年之前 |
Tri Dao