Training in progress, step 14500, checkpoint
Browse files
checkpoint-14500/config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/clip-vit-large-patch14-336",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"CLIPModel"
|
| 5 |
+
],
|
| 6 |
+
"initializer_factor": 1.0,
|
| 7 |
+
"logit_scale_init_value": 2.6592,
|
| 8 |
+
"model_type": "clip",
|
| 9 |
+
"projection_dim": 768,
|
| 10 |
+
"text_config": {
|
| 11 |
+
"dropout": 0.0,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"model_type": "clip_text_model",
|
| 15 |
+
"num_attention_heads": 12,
|
| 16 |
+
"projection_dim": 768
|
| 17 |
+
},
|
| 18 |
+
"torch_dtype": "float32",
|
| 19 |
+
"transformers_version": "4.45.0.dev0",
|
| 20 |
+
"vision_config": {
|
| 21 |
+
"dropout": 0.0,
|
| 22 |
+
"hidden_size": 1024,
|
| 23 |
+
"image_size": 336,
|
| 24 |
+
"intermediate_size": 4096,
|
| 25 |
+
"model_type": "clip_vision_model",
|
| 26 |
+
"num_attention_heads": 16,
|
| 27 |
+
"num_hidden_layers": 24,
|
| 28 |
+
"patch_size": 14,
|
| 29 |
+
"projection_dim": 768
|
| 30 |
+
}
|
| 31 |
+
}
|
checkpoint-14500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c35a5f3f124e7919ef8c7a1de1b107f5e73ff0bd4229883b88c499421142a372
|
| 3 |
+
size 1711848436
|
checkpoint-14500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a738147ae0691ba22be58017199e44418f3cbaf071123bb914274687c745933d
|
| 3 |
+
size 3424043887
|
checkpoint-14500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48472cd9b18a7015cc65e39abf51ba2863d4d4291be21afb626d309680ef0875
|
| 3 |
+
size 14503
|
checkpoint-14500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b813ca94dda94b3d4ea78de0d7929047e86d3b25cbaf2aab0b763206d69c7f40
|
| 3 |
+
size 623
|
checkpoint-14500/trainer_state.json
ADDED
|
@@ -0,0 +1,468 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.26187002658844,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l18-l/checkpoint-14500",
|
| 4 |
+
"epoch": 1.5465017064846416,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 14500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05332764505119454,
|
| 13 |
+
"grad_norm": 186.70851135253906,
|
| 14 |
+
"learning_rate": 9.866680887372013e-09,
|
| 15 |
+
"loss": 0.4804,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05332764505119454,
|
| 20 |
+
"eval_loss": 1.4967491626739502,
|
| 21 |
+
"eval_runtime": 60.9128,
|
| 22 |
+
"eval_samples_per_second": 16.204,
|
| 23 |
+
"eval_steps_per_second": 2.036,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.10665529010238908,
|
| 28 |
+
"grad_norm": 86.50623321533203,
|
| 29 |
+
"learning_rate": 9.733361774744027e-09,
|
| 30 |
+
"loss": 0.4928,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.10665529010238908,
|
| 35 |
+
"eval_loss": 1.482089877128601,
|
| 36 |
+
"eval_runtime": 62.7665,
|
| 37 |
+
"eval_samples_per_second": 15.725,
|
| 38 |
+
"eval_steps_per_second": 1.976,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.1599829351535836,
|
| 43 |
+
"grad_norm": 603.98974609375,
|
| 44 |
+
"learning_rate": 9.600042662116041e-09,
|
| 45 |
+
"loss": 0.4325,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.1599829351535836,
|
| 50 |
+
"eval_loss": 1.4686362743377686,
|
| 51 |
+
"eval_runtime": 61.6915,
|
| 52 |
+
"eval_samples_per_second": 15.999,
|
| 53 |
+
"eval_steps_per_second": 2.01,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.21331058020477817,
|
| 58 |
+
"grad_norm": 13.387365341186523,
|
| 59 |
+
"learning_rate": 9.466723549488055e-09,
|
| 60 |
+
"loss": 0.3964,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.21331058020477817,
|
| 65 |
+
"eval_loss": 1.4561632871627808,
|
| 66 |
+
"eval_runtime": 61.8331,
|
| 67 |
+
"eval_samples_per_second": 15.962,
|
| 68 |
+
"eval_steps_per_second": 2.005,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.2666382252559727,
|
| 73 |
+
"grad_norm": 322.0577392578125,
|
| 74 |
+
"learning_rate": 9.333404436860068e-09,
|
| 75 |
+
"loss": 0.4294,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.2666382252559727,
|
| 80 |
+
"eval_loss": 1.445016622543335,
|
| 81 |
+
"eval_runtime": 63.1527,
|
| 82 |
+
"eval_samples_per_second": 15.629,
|
| 83 |
+
"eval_steps_per_second": 1.963,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.3199658703071672,
|
| 88 |
+
"grad_norm": 0.010774992406368256,
|
| 89 |
+
"learning_rate": 9.200085324232083e-09,
|
| 90 |
+
"loss": 0.3954,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.3199658703071672,
|
| 95 |
+
"eval_loss": 1.4342260360717773,
|
| 96 |
+
"eval_runtime": 63.3562,
|
| 97 |
+
"eval_samples_per_second": 15.579,
|
| 98 |
+
"eval_steps_per_second": 1.957,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.37329351535836175,
|
| 103 |
+
"grad_norm": 42.48988723754883,
|
| 104 |
+
"learning_rate": 9.066766211604096e-09,
|
| 105 |
+
"loss": 0.4116,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.37329351535836175,
|
| 110 |
+
"eval_loss": 1.4254305362701416,
|
| 111 |
+
"eval_runtime": 63.1723,
|
| 112 |
+
"eval_samples_per_second": 15.624,
|
| 113 |
+
"eval_steps_per_second": 1.963,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.42662116040955633,
|
| 118 |
+
"grad_norm": 6.53631067276001,
|
| 119 |
+
"learning_rate": 8.933447098976109e-09,
|
| 120 |
+
"loss": 0.3366,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.42662116040955633,
|
| 125 |
+
"eval_loss": 1.4154412746429443,
|
| 126 |
+
"eval_runtime": 63.1863,
|
| 127 |
+
"eval_samples_per_second": 15.62,
|
| 128 |
+
"eval_steps_per_second": 1.962,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.47994880546075086,
|
| 133 |
+
"grad_norm": 219.85145568847656,
|
| 134 |
+
"learning_rate": 8.800127986348124e-09,
|
| 135 |
+
"loss": 0.3941,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.47994880546075086,
|
| 140 |
+
"eval_loss": 1.406192421913147,
|
| 141 |
+
"eval_runtime": 63.3732,
|
| 142 |
+
"eval_samples_per_second": 15.574,
|
| 143 |
+
"eval_steps_per_second": 1.957,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.5332764505119454,
|
| 148 |
+
"grad_norm": 0.5011767148971558,
|
| 149 |
+
"learning_rate": 8.666808873720137e-09,
|
| 150 |
+
"loss": 0.4351,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.5332764505119454,
|
| 155 |
+
"eval_loss": 1.3958221673965454,
|
| 156 |
+
"eval_runtime": 61.9375,
|
| 157 |
+
"eval_samples_per_second": 15.935,
|
| 158 |
+
"eval_steps_per_second": 2.002,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.58660409556314,
|
| 163 |
+
"grad_norm": 1.465696096420288,
|
| 164 |
+
"learning_rate": 8.53348976109215e-09,
|
| 165 |
+
"loss": 0.4269,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.58660409556314,
|
| 170 |
+
"eval_loss": 1.3861993551254272,
|
| 171 |
+
"eval_runtime": 62.145,
|
| 172 |
+
"eval_samples_per_second": 15.882,
|
| 173 |
+
"eval_steps_per_second": 1.995,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.6399317406143344,
|
| 178 |
+
"grad_norm": 514.3063354492188,
|
| 179 |
+
"learning_rate": 8.400170648464164e-09,
|
| 180 |
+
"loss": 0.3764,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.6399317406143344,
|
| 185 |
+
"eval_loss": 1.377150058746338,
|
| 186 |
+
"eval_runtime": 63.2598,
|
| 187 |
+
"eval_samples_per_second": 15.602,
|
| 188 |
+
"eval_steps_per_second": 1.96,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.693259385665529,
|
| 193 |
+
"grad_norm": 267.90301513671875,
|
| 194 |
+
"learning_rate": 8.266851535836177e-09,
|
| 195 |
+
"loss": 0.3354,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.693259385665529,
|
| 200 |
+
"eval_loss": 1.3687273263931274,
|
| 201 |
+
"eval_runtime": 61.8197,
|
| 202 |
+
"eval_samples_per_second": 15.966,
|
| 203 |
+
"eval_steps_per_second": 2.006,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.7465870307167235,
|
| 208 |
+
"grad_norm": 0.4357345700263977,
|
| 209 |
+
"learning_rate": 8.133532423208192e-09,
|
| 210 |
+
"loss": 0.3666,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.7465870307167235,
|
| 215 |
+
"eval_loss": 1.360427975654602,
|
| 216 |
+
"eval_runtime": 61.8202,
|
| 217 |
+
"eval_samples_per_second": 15.966,
|
| 218 |
+
"eval_steps_per_second": 2.006,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7999146757679181,
|
| 223 |
+
"grad_norm": 151.71505737304688,
|
| 224 |
+
"learning_rate": 8.000213310580205e-09,
|
| 225 |
+
"loss": 0.3991,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7999146757679181,
|
| 230 |
+
"eval_loss": 1.351379156112671,
|
| 231 |
+
"eval_runtime": 62.9768,
|
| 232 |
+
"eval_samples_per_second": 15.672,
|
| 233 |
+
"eval_steps_per_second": 1.969,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.8532423208191127,
|
| 238 |
+
"grad_norm": 467.88653564453125,
|
| 239 |
+
"learning_rate": 7.86689419795222e-09,
|
| 240 |
+
"loss": 0.4028,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.8532423208191127,
|
| 245 |
+
"eval_loss": 1.3436161279678345,
|
| 246 |
+
"eval_runtime": 62.1794,
|
| 247 |
+
"eval_samples_per_second": 15.873,
|
| 248 |
+
"eval_steps_per_second": 1.994,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.9065699658703071,
|
| 253 |
+
"grad_norm": 186.75767517089844,
|
| 254 |
+
"learning_rate": 7.733575085324233e-09,
|
| 255 |
+
"loss": 0.3412,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.9065699658703071,
|
| 260 |
+
"eval_loss": 1.3365025520324707,
|
| 261 |
+
"eval_runtime": 62.3614,
|
| 262 |
+
"eval_samples_per_second": 15.827,
|
| 263 |
+
"eval_steps_per_second": 1.988,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.9598976109215017,
|
| 268 |
+
"grad_norm": 62.3108024597168,
|
| 269 |
+
"learning_rate": 7.600255972696246e-09,
|
| 270 |
+
"loss": 0.3685,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.9598976109215017,
|
| 275 |
+
"eval_loss": 1.3287090063095093,
|
| 276 |
+
"eval_runtime": 63.3854,
|
| 277 |
+
"eval_samples_per_second": 15.571,
|
| 278 |
+
"eval_steps_per_second": 1.956,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.0132252559726962,
|
| 283 |
+
"grad_norm": 537.1036987304688,
|
| 284 |
+
"learning_rate": 7.46693686006826e-09,
|
| 285 |
+
"loss": 0.3502,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.0132252559726962,
|
| 290 |
+
"eval_loss": 1.3230267763137817,
|
| 291 |
+
"eval_runtime": 61.9741,
|
| 292 |
+
"eval_samples_per_second": 15.926,
|
| 293 |
+
"eval_steps_per_second": 2.001,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.0665529010238908,
|
| 298 |
+
"grad_norm": 433.3678283691406,
|
| 299 |
+
"learning_rate": 7.333617747440273e-09,
|
| 300 |
+
"loss": 0.3423,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.0665529010238908,
|
| 305 |
+
"eval_loss": 1.3153482675552368,
|
| 306 |
+
"eval_runtime": 63.108,
|
| 307 |
+
"eval_samples_per_second": 15.64,
|
| 308 |
+
"eval_steps_per_second": 1.965,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.1198805460750854,
|
| 313 |
+
"grad_norm": 16.9348087310791,
|
| 314 |
+
"learning_rate": 7.2002986348122865e-09,
|
| 315 |
+
"loss": 0.3562,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.1198805460750854,
|
| 320 |
+
"eval_loss": 1.307684063911438,
|
| 321 |
+
"eval_runtime": 62.1177,
|
| 322 |
+
"eval_samples_per_second": 15.889,
|
| 323 |
+
"eval_steps_per_second": 1.996,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.17320819112628,
|
| 328 |
+
"grad_norm": 333.85516357421875,
|
| 329 |
+
"learning_rate": 7.0669795221843005e-09,
|
| 330 |
+
"loss": 0.34,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.17320819112628,
|
| 335 |
+
"eval_loss": 1.3010845184326172,
|
| 336 |
+
"eval_runtime": 63.2514,
|
| 337 |
+
"eval_samples_per_second": 15.604,
|
| 338 |
+
"eval_steps_per_second": 1.96,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.2265358361774745,
|
| 343 |
+
"grad_norm": 0.8008983731269836,
|
| 344 |
+
"learning_rate": 6.933660409556314e-09,
|
| 345 |
+
"loss": 0.3681,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.2265358361774745,
|
| 350 |
+
"eval_loss": 1.2945556640625,
|
| 351 |
+
"eval_runtime": 62.1026,
|
| 352 |
+
"eval_samples_per_second": 15.893,
|
| 353 |
+
"eval_steps_per_second": 1.997,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.2798634812286689,
|
| 358 |
+
"grad_norm": 61.41884994506836,
|
| 359 |
+
"learning_rate": 6.800341296928328e-09,
|
| 360 |
+
"loss": 0.304,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.2798634812286689,
|
| 365 |
+
"eval_loss": 1.2891267538070679,
|
| 366 |
+
"eval_runtime": 63.0412,
|
| 367 |
+
"eval_samples_per_second": 15.656,
|
| 368 |
+
"eval_steps_per_second": 1.967,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.3331911262798635,
|
| 373 |
+
"grad_norm": 0.1409968137741089,
|
| 374 |
+
"learning_rate": 6.667022184300341e-09,
|
| 375 |
+
"loss": 0.3365,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.3331911262798635,
|
| 380 |
+
"eval_loss": 1.2835596799850464,
|
| 381 |
+
"eval_runtime": 62.1735,
|
| 382 |
+
"eval_samples_per_second": 15.875,
|
| 383 |
+
"eval_steps_per_second": 1.994,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.386518771331058,
|
| 388 |
+
"grad_norm": 12.641498565673828,
|
| 389 |
+
"learning_rate": 6.533703071672356e-09,
|
| 390 |
+
"loss": 0.323,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.386518771331058,
|
| 395 |
+
"eval_loss": 1.2784309387207031,
|
| 396 |
+
"eval_runtime": 63.1295,
|
| 397 |
+
"eval_samples_per_second": 15.635,
|
| 398 |
+
"eval_steps_per_second": 1.964,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.4398464163822526,
|
| 403 |
+
"grad_norm": 44.88418960571289,
|
| 404 |
+
"learning_rate": 6.400383959044369e-09,
|
| 405 |
+
"loss": 0.3823,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.4398464163822526,
|
| 410 |
+
"eval_loss": 1.2729544639587402,
|
| 411 |
+
"eval_runtime": 62.2234,
|
| 412 |
+
"eval_samples_per_second": 15.862,
|
| 413 |
+
"eval_steps_per_second": 1.993,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.493174061433447,
|
| 418 |
+
"grad_norm": 331.7489013671875,
|
| 419 |
+
"learning_rate": 6.267064846416382e-09,
|
| 420 |
+
"loss": 0.2957,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.493174061433447,
|
| 425 |
+
"eval_loss": 1.267383098602295,
|
| 426 |
+
"eval_runtime": 63.223,
|
| 427 |
+
"eval_samples_per_second": 15.611,
|
| 428 |
+
"eval_steps_per_second": 1.961,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.5465017064846416,
|
| 433 |
+
"grad_norm": 144.97119140625,
|
| 434 |
+
"learning_rate": 6.133745733788396e-09,
|
| 435 |
+
"loss": 0.3459,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.5465017064846416,
|
| 440 |
+
"eval_loss": 1.26187002658844,
|
| 441 |
+
"eval_runtime": 62.0237,
|
| 442 |
+
"eval_samples_per_second": 15.913,
|
| 443 |
+
"eval_steps_per_second": 1.999,
|
| 444 |
+
"step": 14500
|
| 445 |
+
}
|
| 446 |
+
],
|
| 447 |
+
"logging_steps": 500,
|
| 448 |
+
"max_steps": 37504,
|
| 449 |
+
"num_input_tokens_seen": 0,
|
| 450 |
+
"num_train_epochs": 4,
|
| 451 |
+
"save_steps": 500,
|
| 452 |
+
"stateful_callbacks": {
|
| 453 |
+
"TrainerControl": {
|
| 454 |
+
"args": {
|
| 455 |
+
"should_epoch_stop": false,
|
| 456 |
+
"should_evaluate": false,
|
| 457 |
+
"should_log": false,
|
| 458 |
+
"should_save": true,
|
| 459 |
+
"should_training_stop": false
|
| 460 |
+
},
|
| 461 |
+
"attributes": {}
|
| 462 |
+
}
|
| 463 |
+
},
|
| 464 |
+
"total_flos": 5217215609972520.0,
|
| 465 |
+
"train_batch_size": 2,
|
| 466 |
+
"trial_name": null,
|
| 467 |
+
"trial_params": null
|
| 468 |
+
}
|
checkpoint-14500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf62d80c5182e7768a9c6f2a2a23a186cdc24c25dbb6c6b5e997201f7ab22576
|
| 3 |
+
size 4847
|