End of training
Browse files- coreconfig.json +26 -0
- logs/events.out.tfevents.1735394245.gna4000.586124.0 +3 -0
- nvidia_smi_early.log +21 -0
- trainer_state.json +484 -0
- training_args.json +147 -0
coreconfig.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"out_name": "mt5-small-gigatrue",
|
| 3 |
+
"remove_layers": false,
|
| 4 |
+
"model_archetype": "mt5",
|
| 5 |
+
"model_name": "google/mt5-small",
|
| 6 |
+
"tokenizer_name": "google/mt5-small",
|
| 7 |
+
"model_torch_dtype": "bfloat16",
|
| 8 |
+
"dataset_lang": "en",
|
| 9 |
+
"dataset_name": "Plasmoxy/gigatrue",
|
| 10 |
+
"dataset_tokenized_cache_name": "gigatrue_tokenized_mt5_110-35",
|
| 11 |
+
"use_half_val_dataset": true,
|
| 12 |
+
"max_input_length": 110,
|
| 13 |
+
"max_target_length": 35,
|
| 14 |
+
"batch_size": 128,
|
| 15 |
+
"learning_rate": 0.0003,
|
| 16 |
+
"num_train_epochs": 3,
|
| 17 |
+
"pkg_versions": {
|
| 18 |
+
"optimum": "1.23.3",
|
| 19 |
+
"transformers": "4.45.2",
|
| 20 |
+
"openvino": "2024.6.0",
|
| 21 |
+
"nncf": "2.14.1",
|
| 22 |
+
"torch": "2.5.1",
|
| 23 |
+
"datasets": "3.2.0",
|
| 24 |
+
"peft": "0.13.3.dev0"
|
| 25 |
+
}
|
| 26 |
+
}
|
logs/events.out.tfevents.1735394245.gna4000.586124.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ec8f5efb110c259a3cc7320f7312621a9115f0bff389f9ddb85c817d15085e9
|
| 3 |
+
size 20022
|
nvidia_smi_early.log
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Sat Dec 28 14:57:55 2024
|
| 2 |
+
+-----------------------------------------------------------------------------------------+
|
| 3 |
+
| NVIDIA-SMI 550.127.08 Driver Version: 550.127.08 CUDA Version: 12.4 |
|
| 4 |
+
|-----------------------------------------+------------------------+----------------------+
|
| 5 |
+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
|
| 6 |
+
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
|
| 7 |
+
| | | MIG M. |
|
| 8 |
+
|=========================================+========================+======================|
|
| 9 |
+
| 0 NVIDIA RTX A4000 Off | 00000000:01:00.0 Off | Off |
|
| 10 |
+
| 54% 76C P2 135W / 140W | 15525MiB / 16376MiB | 99% Default |
|
| 11 |
+
| | | N/A |
|
| 12 |
+
+-----------------------------------------+------------------------+----------------------+
|
| 13 |
+
|
| 14 |
+
+-----------------------------------------------------------------------------------------+
|
| 15 |
+
| Processes: |
|
| 16 |
+
| GPU GI CI PID Type Process name GPU Memory |
|
| 17 |
+
| ID ID Usage |
|
| 18 |
+
|=========================================================================================|
|
| 19 |
+
| 0 N/A N/A 76833 G /usr/lib/xorg/Xorg 4MiB |
|
| 20 |
+
| 0 N/A N/A 586124 C ...ik/miniforge3/envs/alpha/bin/python 15510MiB |
|
| 21 |
+
+-----------------------------------------------------------------------------------------+
|
trainer_state.json
ADDED
|
@@ -0,0 +1,484 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 3000,
|
| 6 |
+
"global_step": 88686,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 3.382721060821325e-05,
|
| 13 |
+
"grad_norm": 1352.0,
|
| 14 |
+
"learning_rate": 0.00029999661727893914,
|
| 15 |
+
"loss": 22.625,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.10148163182463973,
|
| 20 |
+
"grad_norm": 1.4296875,
|
| 21 |
+
"learning_rate": 0.000289851836817536,
|
| 22 |
+
"loss": 3.5832,
|
| 23 |
+
"step": 3000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.10148163182463973,
|
| 27 |
+
"eval_loss": 2.45703125,
|
| 28 |
+
"eval_runtime": 96.0596,
|
| 29 |
+
"eval_samples_per_second": 982.775,
|
| 30 |
+
"eval_steps_per_second": 7.683,
|
| 31 |
+
"step": 3000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.20296326364927947,
|
| 35 |
+
"grad_norm": 1.375,
|
| 36 |
+
"learning_rate": 0.000279703673635072,
|
| 37 |
+
"loss": 2.9906,
|
| 38 |
+
"step": 6000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.20296326364927947,
|
| 42 |
+
"eval_loss": 2.388899564743042,
|
| 43 |
+
"eval_runtime": 95.3468,
|
| 44 |
+
"eval_samples_per_second": 990.122,
|
| 45 |
+
"eval_steps_per_second": 7.74,
|
| 46 |
+
"step": 6000
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.30444489547391923,
|
| 50 |
+
"grad_norm": 1.3125,
|
| 51 |
+
"learning_rate": 0.00026955551045260807,
|
| 52 |
+
"loss": 2.9367,
|
| 53 |
+
"step": 9000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.30444489547391923,
|
| 57 |
+
"eval_loss": 2.3747141361236572,
|
| 58 |
+
"eval_runtime": 95.1929,
|
| 59 |
+
"eval_samples_per_second": 991.723,
|
| 60 |
+
"eval_steps_per_second": 7.753,
|
| 61 |
+
"step": 9000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.40592652729855894,
|
| 65 |
+
"grad_norm": 1.421875,
|
| 66 |
+
"learning_rate": 0.0002594073472701441,
|
| 67 |
+
"loss": 2.9115,
|
| 68 |
+
"step": 12000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.40592652729855894,
|
| 72 |
+
"eval_loss": 2.3593432903289795,
|
| 73 |
+
"eval_runtime": 95.2533,
|
| 74 |
+
"eval_samples_per_second": 991.095,
|
| 75 |
+
"eval_steps_per_second": 7.748,
|
| 76 |
+
"step": 12000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.5074081591231987,
|
| 80 |
+
"grad_norm": 1.3828125,
|
| 81 |
+
"learning_rate": 0.0002492591840876801,
|
| 82 |
+
"loss": 2.891,
|
| 83 |
+
"step": 15000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.5074081591231987,
|
| 87 |
+
"eval_loss": 2.3490748405456543,
|
| 88 |
+
"eval_runtime": 95.1766,
|
| 89 |
+
"eval_samples_per_second": 991.893,
|
| 90 |
+
"eval_steps_per_second": 7.754,
|
| 91 |
+
"step": 15000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.6088897909478385,
|
| 95 |
+
"grad_norm": 1.3671875,
|
| 96 |
+
"learning_rate": 0.0002391110209052161,
|
| 97 |
+
"loss": 2.8835,
|
| 98 |
+
"step": 18000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.6088897909478385,
|
| 102 |
+
"eval_loss": 2.3357362747192383,
|
| 103 |
+
"eval_runtime": 95.2104,
|
| 104 |
+
"eval_samples_per_second": 991.541,
|
| 105 |
+
"eval_steps_per_second": 7.751,
|
| 106 |
+
"step": 18000
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.7103714227724782,
|
| 110 |
+
"grad_norm": 1.40625,
|
| 111 |
+
"learning_rate": 0.00022896285772275215,
|
| 112 |
+
"loss": 2.8777,
|
| 113 |
+
"step": 21000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.7103714227724782,
|
| 117 |
+
"eval_loss": 2.33984375,
|
| 118 |
+
"eval_runtime": 95.1033,
|
| 119 |
+
"eval_samples_per_second": 992.658,
|
| 120 |
+
"eval_steps_per_second": 7.76,
|
| 121 |
+
"step": 21000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.8118530545971179,
|
| 125 |
+
"grad_norm": 1.4140625,
|
| 126 |
+
"learning_rate": 0.0002188146945402882,
|
| 127 |
+
"loss": 2.8722,
|
| 128 |
+
"step": 24000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.8118530545971179,
|
| 132 |
+
"eval_loss": 2.335789203643799,
|
| 133 |
+
"eval_runtime": 95.1796,
|
| 134 |
+
"eval_samples_per_second": 991.862,
|
| 135 |
+
"eval_steps_per_second": 7.754,
|
| 136 |
+
"step": 24000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.9133346864217576,
|
| 140 |
+
"grad_norm": 1.40625,
|
| 141 |
+
"learning_rate": 0.00020866653135782423,
|
| 142 |
+
"loss": 2.8663,
|
| 143 |
+
"step": 27000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.9133346864217576,
|
| 147 |
+
"eval_loss": 2.3275110721588135,
|
| 148 |
+
"eval_runtime": 95.2734,
|
| 149 |
+
"eval_samples_per_second": 990.885,
|
| 150 |
+
"eval_steps_per_second": 7.746,
|
| 151 |
+
"step": 27000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.0148163182463974,
|
| 155 |
+
"grad_norm": 1.359375,
|
| 156 |
+
"learning_rate": 0.00019851836817536025,
|
| 157 |
+
"loss": 2.8658,
|
| 158 |
+
"step": 30000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.0148163182463974,
|
| 162 |
+
"eval_loss": 2.330390453338623,
|
| 163 |
+
"eval_runtime": 95.2824,
|
| 164 |
+
"eval_samples_per_second": 990.792,
|
| 165 |
+
"eval_steps_per_second": 7.745,
|
| 166 |
+
"step": 30000
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.116297950071037,
|
| 170 |
+
"grad_norm": 1.375,
|
| 171 |
+
"learning_rate": 0.0001883702049928963,
|
| 172 |
+
"loss": 2.8623,
|
| 173 |
+
"step": 33000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.116297950071037,
|
| 177 |
+
"eval_loss": 2.3300411701202393,
|
| 178 |
+
"eval_runtime": 95.3514,
|
| 179 |
+
"eval_samples_per_second": 990.074,
|
| 180 |
+
"eval_steps_per_second": 7.74,
|
| 181 |
+
"step": 33000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.217779581895677,
|
| 185 |
+
"grad_norm": 1.40625,
|
| 186 |
+
"learning_rate": 0.0001782220418104323,
|
| 187 |
+
"loss": 2.8579,
|
| 188 |
+
"step": 36000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.217779581895677,
|
| 192 |
+
"eval_loss": 2.3285059928894043,
|
| 193 |
+
"eval_runtime": 95.288,
|
| 194 |
+
"eval_samples_per_second": 990.734,
|
| 195 |
+
"eval_steps_per_second": 7.745,
|
| 196 |
+
"step": 36000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.3192612137203166,
|
| 200 |
+
"grad_norm": 1.9140625,
|
| 201 |
+
"learning_rate": 0.00016807387862796832,
|
| 202 |
+
"loss": 2.857,
|
| 203 |
+
"step": 39000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.3192612137203166,
|
| 207 |
+
"eval_loss": 2.3231706619262695,
|
| 208 |
+
"eval_runtime": 95.3264,
|
| 209 |
+
"eval_samples_per_second": 990.334,
|
| 210 |
+
"eval_steps_per_second": 7.742,
|
| 211 |
+
"step": 39000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.4207428455449564,
|
| 215 |
+
"grad_norm": 1.359375,
|
| 216 |
+
"learning_rate": 0.00015792571544550436,
|
| 217 |
+
"loss": 2.8552,
|
| 218 |
+
"step": 42000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.4207428455449564,
|
| 222 |
+
"eval_loss": 2.3225038051605225,
|
| 223 |
+
"eval_runtime": 95.2774,
|
| 224 |
+
"eval_samples_per_second": 990.843,
|
| 225 |
+
"eval_steps_per_second": 7.746,
|
| 226 |
+
"step": 42000
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.522224477369596,
|
| 230 |
+
"grad_norm": 1.40625,
|
| 231 |
+
"learning_rate": 0.00014777755226304037,
|
| 232 |
+
"loss": 2.8548,
|
| 233 |
+
"step": 45000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.522224477369596,
|
| 237 |
+
"eval_loss": 2.3205666542053223,
|
| 238 |
+
"eval_runtime": 95.4014,
|
| 239 |
+
"eval_samples_per_second": 989.556,
|
| 240 |
+
"eval_steps_per_second": 7.736,
|
| 241 |
+
"step": 45000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.6237061091942357,
|
| 245 |
+
"grad_norm": 1.4375,
|
| 246 |
+
"learning_rate": 0.0001376293890805764,
|
| 247 |
+
"loss": 2.8518,
|
| 248 |
+
"step": 48000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.6237061091942357,
|
| 252 |
+
"eval_loss": 2.324282169342041,
|
| 253 |
+
"eval_runtime": 95.4158,
|
| 254 |
+
"eval_samples_per_second": 989.407,
|
| 255 |
+
"eval_steps_per_second": 7.735,
|
| 256 |
+
"step": 48000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.7251877410188756,
|
| 260 |
+
"grad_norm": 1.7578125,
|
| 261 |
+
"learning_rate": 0.00012748122589811243,
|
| 262 |
+
"loss": 2.8539,
|
| 263 |
+
"step": 51000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.7251877410188756,
|
| 267 |
+
"eval_loss": 2.3227896690368652,
|
| 268 |
+
"eval_runtime": 95.374,
|
| 269 |
+
"eval_samples_per_second": 989.84,
|
| 270 |
+
"eval_steps_per_second": 7.738,
|
| 271 |
+
"step": 51000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.8266693728435153,
|
| 275 |
+
"grad_norm": 1.34375,
|
| 276 |
+
"learning_rate": 0.00011733306271564845,
|
| 277 |
+
"loss": 2.8483,
|
| 278 |
+
"step": 54000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.8266693728435153,
|
| 282 |
+
"eval_loss": 2.3198044300079346,
|
| 283 |
+
"eval_runtime": 95.1153,
|
| 284 |
+
"eval_samples_per_second": 992.532,
|
| 285 |
+
"eval_steps_per_second": 7.759,
|
| 286 |
+
"step": 54000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.928151004668155,
|
| 290 |
+
"grad_norm": 1.3359375,
|
| 291 |
+
"learning_rate": 0.00010718489953318448,
|
| 292 |
+
"loss": 2.8512,
|
| 293 |
+
"step": 57000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.928151004668155,
|
| 297 |
+
"eval_loss": 2.321180582046509,
|
| 298 |
+
"eval_runtime": 95.1519,
|
| 299 |
+
"eval_samples_per_second": 992.15,
|
| 300 |
+
"eval_steps_per_second": 7.756,
|
| 301 |
+
"step": 57000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 2.029632636492795,
|
| 305 |
+
"grad_norm": 1.3671875,
|
| 306 |
+
"learning_rate": 9.703673635072052e-05,
|
| 307 |
+
"loss": 2.8515,
|
| 308 |
+
"step": 60000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.029632636492795,
|
| 312 |
+
"eval_loss": 2.3204078674316406,
|
| 313 |
+
"eval_runtime": 95.1797,
|
| 314 |
+
"eval_samples_per_second": 991.861,
|
| 315 |
+
"eval_steps_per_second": 7.754,
|
| 316 |
+
"step": 60000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 2.1311142683174347,
|
| 320 |
+
"grad_norm": 1.390625,
|
| 321 |
+
"learning_rate": 8.688857316825655e-05,
|
| 322 |
+
"loss": 2.8512,
|
| 323 |
+
"step": 63000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.1311142683174347,
|
| 327 |
+
"eval_loss": 2.3204712867736816,
|
| 328 |
+
"eval_runtime": 95.1755,
|
| 329 |
+
"eval_samples_per_second": 991.905,
|
| 330 |
+
"eval_steps_per_second": 7.754,
|
| 331 |
+
"step": 63000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.232595900142074,
|
| 335 |
+
"grad_norm": 1.484375,
|
| 336 |
+
"learning_rate": 7.674040998579256e-05,
|
| 337 |
+
"loss": 2.8492,
|
| 338 |
+
"step": 66000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.232595900142074,
|
| 342 |
+
"eval_loss": 2.3218369483947754,
|
| 343 |
+
"eval_runtime": 95.2213,
|
| 344 |
+
"eval_samples_per_second": 991.428,
|
| 345 |
+
"eval_steps_per_second": 7.75,
|
| 346 |
+
"step": 66000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.334077531966714,
|
| 350 |
+
"grad_norm": 1.4296875,
|
| 351 |
+
"learning_rate": 6.659224680332859e-05,
|
| 352 |
+
"loss": 2.851,
|
| 353 |
+
"step": 69000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 2.334077531966714,
|
| 357 |
+
"eval_loss": 2.3220698833465576,
|
| 358 |
+
"eval_runtime": 95.1826,
|
| 359 |
+
"eval_samples_per_second": 991.831,
|
| 360 |
+
"eval_steps_per_second": 7.754,
|
| 361 |
+
"step": 69000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.435559163791354,
|
| 365 |
+
"grad_norm": 1.46875,
|
| 366 |
+
"learning_rate": 5.644408362086462e-05,
|
| 367 |
+
"loss": 2.8497,
|
| 368 |
+
"step": 72000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.435559163791354,
|
| 372 |
+
"eval_loss": 2.320767641067505,
|
| 373 |
+
"eval_runtime": 95.2504,
|
| 374 |
+
"eval_samples_per_second": 991.125,
|
| 375 |
+
"eval_steps_per_second": 7.748,
|
| 376 |
+
"step": 72000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 2.5370407956159937,
|
| 380 |
+
"grad_norm": 1.46875,
|
| 381 |
+
"learning_rate": 4.629592043840065e-05,
|
| 382 |
+
"loss": 2.848,
|
| 383 |
+
"step": 75000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 2.5370407956159937,
|
| 387 |
+
"eval_loss": 2.3203125,
|
| 388 |
+
"eval_runtime": 95.295,
|
| 389 |
+
"eval_samples_per_second": 990.661,
|
| 390 |
+
"eval_steps_per_second": 7.744,
|
| 391 |
+
"step": 75000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.638522427440633,
|
| 395 |
+
"grad_norm": 1.2265625,
|
| 396 |
+
"learning_rate": 3.614775725593667e-05,
|
| 397 |
+
"loss": 2.852,
|
| 398 |
+
"step": 78000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.638522427440633,
|
| 402 |
+
"eval_loss": 2.319963216781616,
|
| 403 |
+
"eval_runtime": 95.168,
|
| 404 |
+
"eval_samples_per_second": 991.982,
|
| 405 |
+
"eval_steps_per_second": 7.755,
|
| 406 |
+
"step": 78000
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.740004059265273,
|
| 410 |
+
"grad_norm": 1.328125,
|
| 411 |
+
"learning_rate": 2.59995940734727e-05,
|
| 412 |
+
"loss": 2.8483,
|
| 413 |
+
"step": 81000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.740004059265273,
|
| 417 |
+
"eval_loss": 2.321169853210449,
|
| 418 |
+
"eval_runtime": 95.17,
|
| 419 |
+
"eval_samples_per_second": 991.962,
|
| 420 |
+
"eval_steps_per_second": 7.755,
|
| 421 |
+
"step": 81000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.841485691089913,
|
| 425 |
+
"grad_norm": 1.4296875,
|
| 426 |
+
"learning_rate": 1.5851430891008727e-05,
|
| 427 |
+
"loss": 2.85,
|
| 428 |
+
"step": 84000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 2.841485691089913,
|
| 432 |
+
"eval_loss": 2.3206405639648438,
|
| 433 |
+
"eval_runtime": 95.2006,
|
| 434 |
+
"eval_samples_per_second": 991.642,
|
| 435 |
+
"eval_steps_per_second": 7.752,
|
| 436 |
+
"step": 84000
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.9429673229145523,
|
| 440 |
+
"grad_norm": 1.375,
|
| 441 |
+
"learning_rate": 5.703267708544753e-06,
|
| 442 |
+
"loss": 2.8503,
|
| 443 |
+
"step": 87000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 2.9429673229145523,
|
| 447 |
+
"eval_loss": 2.320661783218384,
|
| 448 |
+
"eval_runtime": 95.1542,
|
| 449 |
+
"eval_samples_per_second": 992.127,
|
| 450 |
+
"eval_steps_per_second": 7.756,
|
| 451 |
+
"step": 87000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 3.0,
|
| 455 |
+
"step": 88686,
|
| 456 |
+
"total_flos": 1.3129555583125094e+18,
|
| 457 |
+
"train_loss": 2.891493231738944,
|
| 458 |
+
"train_runtime": 37678.5959,
|
| 459 |
+
"train_samples_per_second": 301.271,
|
| 460 |
+
"train_steps_per_second": 2.354
|
| 461 |
+
}
|
| 462 |
+
],
|
| 463 |
+
"logging_steps": 3000,
|
| 464 |
+
"max_steps": 88686,
|
| 465 |
+
"num_input_tokens_seen": 0,
|
| 466 |
+
"num_train_epochs": 3,
|
| 467 |
+
"save_steps": 3000,
|
| 468 |
+
"stateful_callbacks": {
|
| 469 |
+
"TrainerControl": {
|
| 470 |
+
"args": {
|
| 471 |
+
"should_epoch_stop": false,
|
| 472 |
+
"should_evaluate": false,
|
| 473 |
+
"should_log": false,
|
| 474 |
+
"should_save": true,
|
| 475 |
+
"should_training_stop": true
|
| 476 |
+
},
|
| 477 |
+
"attributes": {}
|
| 478 |
+
}
|
| 479 |
+
},
|
| 480 |
+
"total_flos": 1.3129555583125094e+18,
|
| 481 |
+
"train_batch_size": 128,
|
| 482 |
+
"trial_name": null,
|
| 483 |
+
"trial_params": null
|
| 484 |
+
}
|
training_args.json
ADDED
|
@@ -0,0 +1,147 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"output_dir": "out_models/mt5-small-gigatrue",
|
| 3 |
+
"overwrite_output_dir": false,
|
| 4 |
+
"do_train": false,
|
| 5 |
+
"do_eval": true,
|
| 6 |
+
"do_predict": false,
|
| 7 |
+
"eval_strategy": "steps",
|
| 8 |
+
"prediction_loss_only": false,
|
| 9 |
+
"per_device_train_batch_size": 128,
|
| 10 |
+
"per_device_eval_batch_size": 128,
|
| 11 |
+
"per_gpu_train_batch_size": null,
|
| 12 |
+
"per_gpu_eval_batch_size": null,
|
| 13 |
+
"gradient_accumulation_steps": 1,
|
| 14 |
+
"eval_accumulation_steps": null,
|
| 15 |
+
"eval_delay": 0,
|
| 16 |
+
"torch_empty_cache_steps": null,
|
| 17 |
+
"learning_rate": 0.0003,
|
| 18 |
+
"weight_decay": 0.0,
|
| 19 |
+
"adam_beta1": 0.9,
|
| 20 |
+
"adam_beta2": 0.999,
|
| 21 |
+
"adam_epsilon": 1e-08,
|
| 22 |
+
"max_grad_norm": 1.0,
|
| 23 |
+
"num_train_epochs": 3,
|
| 24 |
+
"max_steps": -1,
|
| 25 |
+
"lr_scheduler_type": "linear",
|
| 26 |
+
"lr_scheduler_kwargs": {},
|
| 27 |
+
"warmup_ratio": 0.0,
|
| 28 |
+
"warmup_steps": 0,
|
| 29 |
+
"log_level": "passive",
|
| 30 |
+
"log_level_replica": "warning",
|
| 31 |
+
"log_on_each_node": true,
|
| 32 |
+
"logging_dir": "out_models/mt5-small-gigatrue/logs",
|
| 33 |
+
"logging_strategy": "steps",
|
| 34 |
+
"logging_first_step": true,
|
| 35 |
+
"logging_steps": 3000,
|
| 36 |
+
"logging_nan_inf_filter": true,
|
| 37 |
+
"save_strategy": "steps",
|
| 38 |
+
"save_steps": 3000,
|
| 39 |
+
"save_total_limit": null,
|
| 40 |
+
"save_safetensors": true,
|
| 41 |
+
"save_on_each_node": false,
|
| 42 |
+
"save_only_model": false,
|
| 43 |
+
"restore_callback_states_from_checkpoint": false,
|
| 44 |
+
"no_cuda": false,
|
| 45 |
+
"use_cpu": false,
|
| 46 |
+
"use_mps_device": false,
|
| 47 |
+
"seed": 42,
|
| 48 |
+
"data_seed": null,
|
| 49 |
+
"jit_mode_eval": false,
|
| 50 |
+
"use_ipex": false,
|
| 51 |
+
"bf16": false,
|
| 52 |
+
"fp16": false,
|
| 53 |
+
"fp16_opt_level": "O1",
|
| 54 |
+
"half_precision_backend": "auto",
|
| 55 |
+
"bf16_full_eval": false,
|
| 56 |
+
"fp16_full_eval": false,
|
| 57 |
+
"tf32": null,
|
| 58 |
+
"local_rank": 0,
|
| 59 |
+
"ddp_backend": null,
|
| 60 |
+
"tpu_num_cores": null,
|
| 61 |
+
"tpu_metrics_debug": false,
|
| 62 |
+
"debug": [],
|
| 63 |
+
"dataloader_drop_last": false,
|
| 64 |
+
"eval_steps": 3000,
|
| 65 |
+
"dataloader_num_workers": 0,
|
| 66 |
+
"dataloader_prefetch_factor": null,
|
| 67 |
+
"past_index": -1,
|
| 68 |
+
"run_name": "out_models/mt5-small-gigatrue",
|
| 69 |
+
"disable_tqdm": false,
|
| 70 |
+
"remove_unused_columns": true,
|
| 71 |
+
"label_names": null,
|
| 72 |
+
"load_best_model_at_end": false,
|
| 73 |
+
"metric_for_best_model": null,
|
| 74 |
+
"greater_is_better": null,
|
| 75 |
+
"ignore_data_skip": false,
|
| 76 |
+
"fsdp": [],
|
| 77 |
+
"fsdp_min_num_params": 0,
|
| 78 |
+
"fsdp_config": {
|
| 79 |
+
"min_num_params": 0,
|
| 80 |
+
"xla": false,
|
| 81 |
+
"xla_fsdp_v2": false,
|
| 82 |
+
"xla_fsdp_grad_ckpt": false
|
| 83 |
+
},
|
| 84 |
+
"fsdp_transformer_layer_cls_to_wrap": null,
|
| 85 |
+
"accelerator_config": "AcceleratorConfig",
|
| 86 |
+
"deepspeed": null,
|
| 87 |
+
"label_smoothing_factor": 0.0,
|
| 88 |
+
"optim": "adamw_torch",
|
| 89 |
+
"optim_args": null,
|
| 90 |
+
"adafactor": false,
|
| 91 |
+
"group_by_length": false,
|
| 92 |
+
"length_column_name": "length",
|
| 93 |
+
"report_to": [
|
| 94 |
+
"tensorboard"
|
| 95 |
+
],
|
| 96 |
+
"ddp_find_unused_parameters": null,
|
| 97 |
+
"ddp_bucket_cap_mb": null,
|
| 98 |
+
"ddp_broadcast_buffers": null,
|
| 99 |
+
"dataloader_pin_memory": true,
|
| 100 |
+
"dataloader_persistent_workers": false,
|
| 101 |
+
"skip_memory_metrics": true,
|
| 102 |
+
"use_legacy_prediction_loop": false,
|
| 103 |
+
"push_to_hub": false,
|
| 104 |
+
"resume_from_checkpoint": null,
|
| 105 |
+
"hub_model_id": null,
|
| 106 |
+
"hub_strategy": "every_save",
|
| 107 |
+
"hub_token": null,
|
| 108 |
+
"hub_private_repo": false,
|
| 109 |
+
"hub_always_push": false,
|
| 110 |
+
"gradient_checkpointing": false,
|
| 111 |
+
"gradient_checkpointing_kwargs": null,
|
| 112 |
+
"include_inputs_for_metrics": false,
|
| 113 |
+
"eval_do_concat_batches": true,
|
| 114 |
+
"fp16_backend": "auto",
|
| 115 |
+
"evaluation_strategy": null,
|
| 116 |
+
"push_to_hub_model_id": null,
|
| 117 |
+
"push_to_hub_organization": null,
|
| 118 |
+
"push_to_hub_token": null,
|
| 119 |
+
"mp_parameters": "",
|
| 120 |
+
"auto_find_batch_size": false,
|
| 121 |
+
"full_determinism": false,
|
| 122 |
+
"torchdynamo": null,
|
| 123 |
+
"ray_scope": "last",
|
| 124 |
+
"ddp_timeout": 1800,
|
| 125 |
+
"torch_compile": false,
|
| 126 |
+
"torch_compile_backend": null,
|
| 127 |
+
"torch_compile_mode": null,
|
| 128 |
+
"dispatch_batches": null,
|
| 129 |
+
"split_batches": null,
|
| 130 |
+
"include_tokens_per_second": false,
|
| 131 |
+
"include_num_input_tokens_seen": false,
|
| 132 |
+
"neftune_noise_alpha": null,
|
| 133 |
+
"optim_target_modules": null,
|
| 134 |
+
"batch_eval_metrics": false,
|
| 135 |
+
"eval_on_start": false,
|
| 136 |
+
"use_liger_kernel": false,
|
| 137 |
+
"eval_use_gather_object": false,
|
| 138 |
+
"sortish_sampler": false,
|
| 139 |
+
"predict_with_generate": false,
|
| 140 |
+
"generation_max_length": null,
|
| 141 |
+
"generation_num_beams": null,
|
| 142 |
+
"generation_config": null,
|
| 143 |
+
"distributed_state": "PartialState",
|
| 144 |
+
"_n_gpu": 1,
|
| 145 |
+
"__cached__setup_devices": "device",
|
| 146 |
+
"deepspeed_plugin": null
|
| 147 |
+
}
|