Training in progress, step 16000, checkpoint
Browse files
checkpoint-16000/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-16000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e1fea127b01a49245ab17922c0011e053c0a24d15c6021f8d5eae8fc0af61b2
|
| 3 |
+
size 1711848436
|
checkpoint-16000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b396d1af2e9ae08a9140379d4b34b6ecc755321922b32fbf3749c79fd73dab1
|
| 3 |
+
size 3424043887
|
checkpoint-16000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f1c45c04b203caa5e64a0fe73a973b302ef17143590e9d4304433b96f734879b
|
| 3 |
+
size 14503
|
checkpoint-16000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b9ea2e28f5e94d8bc5ab21f2de42ec04f31faeb16c6fbb29c56863f89bc0be6a
|
| 3 |
+
size 623
|
checkpoint-16000/trainer_state.json
ADDED
|
@@ -0,0 +1,513 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.9056034088134766,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l55-l/checkpoint-13000",
|
| 4 |
+
"epoch": 2.947135752440597,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 16000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09209799226376865,
|
| 13 |
+
"grad_norm": 0.009597355499863625,
|
| 14 |
+
"learning_rate": 4.846503346227052e-05,
|
| 15 |
+
"loss": 0.7336,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09209799226376865,
|
| 20 |
+
"eval_loss": 2.0780556201934814,
|
| 21 |
+
"eval_runtime": 77.1404,
|
| 22 |
+
"eval_samples_per_second": 15.634,
|
| 23 |
+
"eval_steps_per_second": 1.957,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.1841959845275373,
|
| 28 |
+
"grad_norm": 0.10049372166395187,
|
| 29 |
+
"learning_rate": 4.693006692454105e-05,
|
| 30 |
+
"loss": 0.6953,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.1841959845275373,
|
| 35 |
+
"eval_loss": 2.0782744884490967,
|
| 36 |
+
"eval_runtime": 77.8965,
|
| 37 |
+
"eval_samples_per_second": 15.482,
|
| 38 |
+
"eval_steps_per_second": 1.938,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.27629397679130596,
|
| 43 |
+
"grad_norm": 0.0041180383414030075,
|
| 44 |
+
"learning_rate": 4.539510038681157e-05,
|
| 45 |
+
"loss": 0.6953,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.27629397679130596,
|
| 50 |
+
"eval_loss": 2.077699899673462,
|
| 51 |
+
"eval_runtime": 78.9586,
|
| 52 |
+
"eval_samples_per_second": 15.274,
|
| 53 |
+
"eval_steps_per_second": 1.912,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.3683919690550746,
|
| 58 |
+
"grad_norm": 0.29397013783454895,
|
| 59 |
+
"learning_rate": 4.386013384908209e-05,
|
| 60 |
+
"loss": 0.6932,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.3683919690550746,
|
| 65 |
+
"eval_loss": 2.0721774101257324,
|
| 66 |
+
"eval_runtime": 77.349,
|
| 67 |
+
"eval_samples_per_second": 15.592,
|
| 68 |
+
"eval_steps_per_second": 1.952,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.46048996131884323,
|
| 73 |
+
"grad_norm": 0.9136261343955994,
|
| 74 |
+
"learning_rate": 4.232516731135262e-05,
|
| 75 |
+
"loss": 0.6907,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.46048996131884323,
|
| 80 |
+
"eval_loss": 2.079211711883545,
|
| 81 |
+
"eval_runtime": 79.3204,
|
| 82 |
+
"eval_samples_per_second": 15.204,
|
| 83 |
+
"eval_steps_per_second": 1.904,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.5525879535826119,
|
| 88 |
+
"grad_norm": 4.484213829040527,
|
| 89 |
+
"learning_rate": 4.079020077362314e-05,
|
| 90 |
+
"loss": 0.7043,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.5525879535826119,
|
| 95 |
+
"eval_loss": 2.0204577445983887,
|
| 96 |
+
"eval_runtime": 78.612,
|
| 97 |
+
"eval_samples_per_second": 15.341,
|
| 98 |
+
"eval_steps_per_second": 1.921,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.6446859458463805,
|
| 103 |
+
"grad_norm": 1.1021015644073486,
|
| 104 |
+
"learning_rate": 3.9255234235893664e-05,
|
| 105 |
+
"loss": 0.7078,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.6446859458463805,
|
| 110 |
+
"eval_loss": 2.1256344318389893,
|
| 111 |
+
"eval_runtime": 78.63,
|
| 112 |
+
"eval_samples_per_second": 15.338,
|
| 113 |
+
"eval_steps_per_second": 1.92,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7367839381101492,
|
| 118 |
+
"grad_norm": 16.864044189453125,
|
| 119 |
+
"learning_rate": 3.772026769816418e-05,
|
| 120 |
+
"loss": 0.7034,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7367839381101492,
|
| 125 |
+
"eval_loss": 2.0000314712524414,
|
| 126 |
+
"eval_runtime": 78.6543,
|
| 127 |
+
"eval_samples_per_second": 15.333,
|
| 128 |
+
"eval_steps_per_second": 1.92,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.8288819303739179,
|
| 133 |
+
"grad_norm": 20.071943283081055,
|
| 134 |
+
"learning_rate": 3.6185301160434705e-05,
|
| 135 |
+
"loss": 0.6671,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.8288819303739179,
|
| 140 |
+
"eval_loss": 2.1607539653778076,
|
| 141 |
+
"eval_runtime": 78.6404,
|
| 142 |
+
"eval_samples_per_second": 15.336,
|
| 143 |
+
"eval_steps_per_second": 1.92,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.9209799226376865,
|
| 148 |
+
"grad_norm": 17.817607879638672,
|
| 149 |
+
"learning_rate": 3.4650334622705225e-05,
|
| 150 |
+
"loss": 0.6899,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.9209799226376865,
|
| 155 |
+
"eval_loss": 2.093959331512451,
|
| 156 |
+
"eval_runtime": 78.539,
|
| 157 |
+
"eval_samples_per_second": 15.355,
|
| 158 |
+
"eval_steps_per_second": 1.923,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.013077914901455,
|
| 163 |
+
"grad_norm": 41.0746955871582,
|
| 164 |
+
"learning_rate": 3.311536808497575e-05,
|
| 165 |
+
"loss": 0.6649,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.013077914901455,
|
| 170 |
+
"eval_loss": 1.9776026010513306,
|
| 171 |
+
"eval_runtime": 78.5095,
|
| 172 |
+
"eval_samples_per_second": 15.361,
|
| 173 |
+
"eval_steps_per_second": 1.923,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.1051759071652238,
|
| 178 |
+
"grad_norm": 35.426021575927734,
|
| 179 |
+
"learning_rate": 3.158040154724627e-05,
|
| 180 |
+
"loss": 0.6748,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.1051759071652238,
|
| 185 |
+
"eval_loss": 1.9851746559143066,
|
| 186 |
+
"eval_runtime": 78.5944,
|
| 187 |
+
"eval_samples_per_second": 15.345,
|
| 188 |
+
"eval_steps_per_second": 1.921,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.1972738994289924,
|
| 193 |
+
"grad_norm": 0.8773899078369141,
|
| 194 |
+
"learning_rate": 3.0045435009516792e-05,
|
| 195 |
+
"loss": 0.6456,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.1972738994289924,
|
| 200 |
+
"eval_loss": 1.9981426000595093,
|
| 201 |
+
"eval_runtime": 78.4741,
|
| 202 |
+
"eval_samples_per_second": 15.368,
|
| 203 |
+
"eval_steps_per_second": 1.924,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.289371891692761,
|
| 208 |
+
"grad_norm": 19.7215518951416,
|
| 209 |
+
"learning_rate": 2.8510468471787316e-05,
|
| 210 |
+
"loss": 0.6514,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.289371891692761,
|
| 215 |
+
"eval_loss": 1.9636509418487549,
|
| 216 |
+
"eval_runtime": 78.4704,
|
| 217 |
+
"eval_samples_per_second": 15.369,
|
| 218 |
+
"eval_steps_per_second": 1.924,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.3814698839565298,
|
| 223 |
+
"grad_norm": 5.96924352645874,
|
| 224 |
+
"learning_rate": 2.697550193405784e-05,
|
| 225 |
+
"loss": 0.6179,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.3814698839565298,
|
| 230 |
+
"eval_loss": 2.0034029483795166,
|
| 231 |
+
"eval_runtime": 78.4771,
|
| 232 |
+
"eval_samples_per_second": 15.368,
|
| 233 |
+
"eval_steps_per_second": 1.924,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.4735678762202984,
|
| 238 |
+
"grad_norm": 1.2145048379898071,
|
| 239 |
+
"learning_rate": 2.5440535396328363e-05,
|
| 240 |
+
"loss": 0.6561,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.4735678762202984,
|
| 245 |
+
"eval_loss": 2.2864511013031006,
|
| 246 |
+
"eval_runtime": 78.3606,
|
| 247 |
+
"eval_samples_per_second": 15.39,
|
| 248 |
+
"eval_steps_per_second": 1.927,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.565665868484067,
|
| 253 |
+
"grad_norm": 15.816754341125488,
|
| 254 |
+
"learning_rate": 2.3905568858598883e-05,
|
| 255 |
+
"loss": 0.6328,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.565665868484067,
|
| 260 |
+
"eval_loss": 2.4807703495025635,
|
| 261 |
+
"eval_runtime": 77.0504,
|
| 262 |
+
"eval_samples_per_second": 15.652,
|
| 263 |
+
"eval_steps_per_second": 1.96,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.6577638607478358,
|
| 268 |
+
"grad_norm": 18.753801345825195,
|
| 269 |
+
"learning_rate": 2.2370602320869404e-05,
|
| 270 |
+
"loss": 0.7053,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.6577638607478358,
|
| 275 |
+
"eval_loss": 2.1609902381896973,
|
| 276 |
+
"eval_runtime": 76.5781,
|
| 277 |
+
"eval_samples_per_second": 15.749,
|
| 278 |
+
"eval_steps_per_second": 1.972,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.7498618530116044,
|
| 283 |
+
"grad_norm": 3.3683111667633057,
|
| 284 |
+
"learning_rate": 2.0835635783139927e-05,
|
| 285 |
+
"loss": 0.6533,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.7498618530116044,
|
| 290 |
+
"eval_loss": 1.9845130443572998,
|
| 291 |
+
"eval_runtime": 78.8415,
|
| 292 |
+
"eval_samples_per_second": 15.297,
|
| 293 |
+
"eval_steps_per_second": 1.915,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.841959845275373,
|
| 298 |
+
"grad_norm": 5.0674967765808105,
|
| 299 |
+
"learning_rate": 1.930066924541045e-05,
|
| 300 |
+
"loss": 0.6594,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.841959845275373,
|
| 305 |
+
"eval_loss": 1.9894555807113647,
|
| 306 |
+
"eval_runtime": 77.1087,
|
| 307 |
+
"eval_samples_per_second": 15.64,
|
| 308 |
+
"eval_steps_per_second": 1.958,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.9340578375391417,
|
| 313 |
+
"grad_norm": 45.641239166259766,
|
| 314 |
+
"learning_rate": 1.7765702707680974e-05,
|
| 315 |
+
"loss": 0.6597,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.9340578375391417,
|
| 320 |
+
"eval_loss": 2.0103421211242676,
|
| 321 |
+
"eval_runtime": 77.0894,
|
| 322 |
+
"eval_samples_per_second": 15.644,
|
| 323 |
+
"eval_steps_per_second": 1.959,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.02615582980291,
|
| 328 |
+
"grad_norm": 7.026269435882568,
|
| 329 |
+
"learning_rate": 1.6230736169951498e-05,
|
| 330 |
+
"loss": 0.6435,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.02615582980291,
|
| 335 |
+
"eval_loss": 2.058932304382324,
|
| 336 |
+
"eval_runtime": 78.5759,
|
| 337 |
+
"eval_samples_per_second": 15.348,
|
| 338 |
+
"eval_steps_per_second": 1.922,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.118253822066679,
|
| 343 |
+
"grad_norm": 11.815095901489258,
|
| 344 |
+
"learning_rate": 1.4695769632222017e-05,
|
| 345 |
+
"loss": 0.6404,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.118253822066679,
|
| 350 |
+
"eval_loss": 2.0517008304595947,
|
| 351 |
+
"eval_runtime": 76.9172,
|
| 352 |
+
"eval_samples_per_second": 15.679,
|
| 353 |
+
"eval_steps_per_second": 1.963,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.2103518143304477,
|
| 358 |
+
"grad_norm": 3.2536027431488037,
|
| 359 |
+
"learning_rate": 1.316080309449254e-05,
|
| 360 |
+
"loss": 0.6348,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.2103518143304477,
|
| 365 |
+
"eval_loss": 1.9749338626861572,
|
| 366 |
+
"eval_runtime": 78.0449,
|
| 367 |
+
"eval_samples_per_second": 15.453,
|
| 368 |
+
"eval_steps_per_second": 1.935,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.302449806594216,
|
| 373 |
+
"grad_norm": 12.233312606811523,
|
| 374 |
+
"learning_rate": 1.1625836556763064e-05,
|
| 375 |
+
"loss": 0.6545,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 2.302449806594216,
|
| 380 |
+
"eval_loss": 1.9680836200714111,
|
| 381 |
+
"eval_runtime": 76.9869,
|
| 382 |
+
"eval_samples_per_second": 15.665,
|
| 383 |
+
"eval_steps_per_second": 1.961,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.394547798857985,
|
| 388 |
+
"grad_norm": 16.69487953186035,
|
| 389 |
+
"learning_rate": 1.0090870019033586e-05,
|
| 390 |
+
"loss": 0.6323,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.394547798857985,
|
| 395 |
+
"eval_loss": 1.9056034088134766,
|
| 396 |
+
"eval_runtime": 78.3412,
|
| 397 |
+
"eval_samples_per_second": 15.394,
|
| 398 |
+
"eval_steps_per_second": 1.927,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.4866457911217537,
|
| 403 |
+
"grad_norm": 11.504305839538574,
|
| 404 |
+
"learning_rate": 8.555903481304108e-06,
|
| 405 |
+
"loss": 0.6045,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.4866457911217537,
|
| 410 |
+
"eval_loss": 1.9470211267471313,
|
| 411 |
+
"eval_runtime": 77.1033,
|
| 412 |
+
"eval_samples_per_second": 15.641,
|
| 413 |
+
"eval_steps_per_second": 1.958,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.578743783385522,
|
| 418 |
+
"grad_norm": 188.8672332763672,
|
| 419 |
+
"learning_rate": 7.0209369435746295e-06,
|
| 420 |
+
"loss": 0.6492,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.578743783385522,
|
| 425 |
+
"eval_loss": 2.017714023590088,
|
| 426 |
+
"eval_runtime": 77.6236,
|
| 427 |
+
"eval_samples_per_second": 15.537,
|
| 428 |
+
"eval_steps_per_second": 1.945,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.670841775649291,
|
| 433 |
+
"grad_norm": 10.520608901977539,
|
| 434 |
+
"learning_rate": 5.485970405845152e-06,
|
| 435 |
+
"loss": 0.6512,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.670841775649291,
|
| 440 |
+
"eval_loss": 1.948345422744751,
|
| 441 |
+
"eval_runtime": 77.8588,
|
| 442 |
+
"eval_samples_per_second": 15.49,
|
| 443 |
+
"eval_steps_per_second": 1.939,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.7629397679130596,
|
| 448 |
+
"grad_norm": 13.767060279846191,
|
| 449 |
+
"learning_rate": 3.951003868115676e-06,
|
| 450 |
+
"loss": 0.6382,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 2.7629397679130596,
|
| 455 |
+
"eval_loss": 1.9499462842941284,
|
| 456 |
+
"eval_runtime": 77.4986,
|
| 457 |
+
"eval_samples_per_second": 15.562,
|
| 458 |
+
"eval_steps_per_second": 1.948,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.8550377601768284,
|
| 463 |
+
"grad_norm": 0.9124027490615845,
|
| 464 |
+
"learning_rate": 2.416037330386198e-06,
|
| 465 |
+
"loss": 0.6203,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.8550377601768284,
|
| 470 |
+
"eval_loss": 1.9528756141662598,
|
| 471 |
+
"eval_runtime": 77.9576,
|
| 472 |
+
"eval_samples_per_second": 15.47,
|
| 473 |
+
"eval_steps_per_second": 1.937,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.947135752440597,
|
| 478 |
+
"grad_norm": 8.062443733215332,
|
| 479 |
+
"learning_rate": 8.810707926567202e-07,
|
| 480 |
+
"loss": 0.6241,
|
| 481 |
+
"step": 16000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 2.947135752440597,
|
| 485 |
+
"eval_loss": 1.981693148612976,
|
| 486 |
+
"eval_runtime": 77.6267,
|
| 487 |
+
"eval_samples_per_second": 15.536,
|
| 488 |
+
"eval_steps_per_second": 1.945,
|
| 489 |
+
"step": 16000
|
| 490 |
+
}
|
| 491 |
+
],
|
| 492 |
+
"logging_steps": 500,
|
| 493 |
+
"max_steps": 16287,
|
| 494 |
+
"num_input_tokens_seen": 0,
|
| 495 |
+
"num_train_epochs": 3,
|
| 496 |
+
"save_steps": 500,
|
| 497 |
+
"stateful_callbacks": {
|
| 498 |
+
"TrainerControl": {
|
| 499 |
+
"args": {
|
| 500 |
+
"should_epoch_stop": false,
|
| 501 |
+
"should_evaluate": false,
|
| 502 |
+
"should_log": false,
|
| 503 |
+
"should_save": true,
|
| 504 |
+
"should_training_stop": false
|
| 505 |
+
},
|
| 506 |
+
"attributes": {}
|
| 507 |
+
}
|
| 508 |
+
},
|
| 509 |
+
"total_flos": 5756493378511260.0,
|
| 510 |
+
"train_batch_size": 2,
|
| 511 |
+
"trial_name": null,
|
| 512 |
+
"trial_params": null
|
| 513 |
+
}
|
checkpoint-16000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9885fd590ddfbffc4f8db1738c1993002dfa8c1603f7992a1cda9b93bbe7d34
|
| 3 |
+
size 4847
|