Training in progress, step 18000, checkpoint
Browse files
checkpoint-18000/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-18000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a76482c8d18b066b778899e684ab7867215262fd41a2aed63b034dd6bc3a0bf
|
| 3 |
+
size 1711848436
|
checkpoint-18000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51a295fc1eb080b63063c5e6d61b7cc0fb6091a60865acf8b4ccc6362c17b7e6
|
| 3 |
+
size 3424043887
|
checkpoint-18000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72b359790c10a2931105aea5a90a1929295d8e08cd2e8572d7169cfaa80c130a
|
| 3 |
+
size 14503
|
checkpoint-18000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0769f24f9bd0ecdb1e071d3481046892a60889b3f81a95ccbb4814e9cfb42561
|
| 3 |
+
size 623
|
checkpoint-18000/trainer_state.json
ADDED
|
@@ -0,0 +1,573 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6002796292304993,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-l/checkpoint-18000",
|
| 4 |
+
"epoch": 2.026342451874367,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 18000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.056287290329843524,
|
| 13 |
+
"grad_norm": 96.92656707763672,
|
| 14 |
+
"learning_rate": 2.9437127096701565e-07,
|
| 15 |
+
"loss": 0.3748,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.056287290329843524,
|
| 20 |
+
"eval_loss": 1.2226382493972778,
|
| 21 |
+
"eval_runtime": 131.2779,
|
| 22 |
+
"eval_samples_per_second": 15.037,
|
| 23 |
+
"eval_steps_per_second": 1.882,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.11257458065968705,
|
| 28 |
+
"grad_norm": 402.31158447265625,
|
| 29 |
+
"learning_rate": 2.887425419340313e-07,
|
| 30 |
+
"loss": 0.3057,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.11257458065968705,
|
| 35 |
+
"eval_loss": 1.0702259540557861,
|
| 36 |
+
"eval_runtime": 131.0109,
|
| 37 |
+
"eval_samples_per_second": 15.067,
|
| 38 |
+
"eval_steps_per_second": 1.885,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.16886187098953057,
|
| 43 |
+
"grad_norm": 515.0911865234375,
|
| 44 |
+
"learning_rate": 2.831138129010469e-07,
|
| 45 |
+
"loss": 0.2239,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.16886187098953057,
|
| 50 |
+
"eval_loss": 0.9957238435745239,
|
| 51 |
+
"eval_runtime": 134.8377,
|
| 52 |
+
"eval_samples_per_second": 14.64,
|
| 53 |
+
"eval_steps_per_second": 1.832,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.2251491613193741,
|
| 58 |
+
"grad_norm": 3.1967105865478516,
|
| 59 |
+
"learning_rate": 2.774850838680626e-07,
|
| 60 |
+
"loss": 0.2229,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.2251491613193741,
|
| 65 |
+
"eval_loss": 0.9504629373550415,
|
| 66 |
+
"eval_runtime": 129.0169,
|
| 67 |
+
"eval_samples_per_second": 15.3,
|
| 68 |
+
"eval_steps_per_second": 1.914,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.2814364516492176,
|
| 73 |
+
"grad_norm": 276.22998046875,
|
| 74 |
+
"learning_rate": 2.718563548350782e-07,
|
| 75 |
+
"loss": 0.2098,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.2814364516492176,
|
| 80 |
+
"eval_loss": 0.9006705284118652,
|
| 81 |
+
"eval_runtime": 129.1816,
|
| 82 |
+
"eval_samples_per_second": 15.281,
|
| 83 |
+
"eval_steps_per_second": 1.912,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.33772374197906113,
|
| 88 |
+
"grad_norm": 1105.317138671875,
|
| 89 |
+
"learning_rate": 2.6622762580209386e-07,
|
| 90 |
+
"loss": 0.1938,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.33772374197906113,
|
| 95 |
+
"eval_loss": 0.8782906532287598,
|
| 96 |
+
"eval_runtime": 128.6387,
|
| 97 |
+
"eval_samples_per_second": 15.345,
|
| 98 |
+
"eval_steps_per_second": 1.92,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.39401103230890466,
|
| 103 |
+
"grad_norm": 0.00043465400813147426,
|
| 104 |
+
"learning_rate": 2.605988967691095e-07,
|
| 105 |
+
"loss": 0.1688,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.39401103230890466,
|
| 110 |
+
"eval_loss": 0.8405746221542358,
|
| 111 |
+
"eval_runtime": 128.6989,
|
| 112 |
+
"eval_samples_per_second": 15.338,
|
| 113 |
+
"eval_steps_per_second": 1.919,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.4502983226387482,
|
| 118 |
+
"grad_norm": 0.027426382526755333,
|
| 119 |
+
"learning_rate": 2.549701677361252e-07,
|
| 120 |
+
"loss": 0.1457,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4502983226387482,
|
| 125 |
+
"eval_loss": 0.813768208026886,
|
| 126 |
+
"eval_runtime": 128.7209,
|
| 127 |
+
"eval_samples_per_second": 15.336,
|
| 128 |
+
"eval_steps_per_second": 1.919,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.5065856129685917,
|
| 133 |
+
"grad_norm": 0.0012935090344399214,
|
| 134 |
+
"learning_rate": 2.4934143870314085e-07,
|
| 135 |
+
"loss": 0.179,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.5065856129685917,
|
| 140 |
+
"eval_loss": 0.7965527176856995,
|
| 141 |
+
"eval_runtime": 128.3969,
|
| 142 |
+
"eval_samples_per_second": 15.374,
|
| 143 |
+
"eval_steps_per_second": 1.924,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.5628729032984352,
|
| 148 |
+
"grad_norm": 0.0017458726651966572,
|
| 149 |
+
"learning_rate": 2.4371270967015646e-07,
|
| 150 |
+
"loss": 0.1224,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.5628729032984352,
|
| 155 |
+
"eval_loss": 0.7788484692573547,
|
| 156 |
+
"eval_runtime": 128.4939,
|
| 157 |
+
"eval_samples_per_second": 15.363,
|
| 158 |
+
"eval_steps_per_second": 1.922,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.6191601936282788,
|
| 163 |
+
"grad_norm": 65.76844787597656,
|
| 164 |
+
"learning_rate": 2.380839806371721e-07,
|
| 165 |
+
"loss": 0.1551,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.6191601936282788,
|
| 170 |
+
"eval_loss": 0.7626588344573975,
|
| 171 |
+
"eval_runtime": 128.7302,
|
| 172 |
+
"eval_samples_per_second": 15.334,
|
| 173 |
+
"eval_steps_per_second": 1.919,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.6754474839581223,
|
| 178 |
+
"grad_norm": 5.014094829559326,
|
| 179 |
+
"learning_rate": 2.3245525160418776e-07,
|
| 180 |
+
"loss": 0.1721,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.6754474839581223,
|
| 185 |
+
"eval_loss": 0.7464810609817505,
|
| 186 |
+
"eval_runtime": 128.749,
|
| 187 |
+
"eval_samples_per_second": 15.332,
|
| 188 |
+
"eval_steps_per_second": 1.918,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.7317347742879657,
|
| 193 |
+
"grad_norm": 0.0022696161177009344,
|
| 194 |
+
"learning_rate": 2.268265225712034e-07,
|
| 195 |
+
"loss": 0.1532,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.7317347742879657,
|
| 200 |
+
"eval_loss": 0.7335842251777649,
|
| 201 |
+
"eval_runtime": 128.8165,
|
| 202 |
+
"eval_samples_per_second": 15.324,
|
| 203 |
+
"eval_steps_per_second": 1.917,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.7880220646178093,
|
| 208 |
+
"grad_norm": 5.508715730684344e-06,
|
| 209 |
+
"learning_rate": 2.2119779353821906e-07,
|
| 210 |
+
"loss": 0.1991,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.7880220646178093,
|
| 215 |
+
"eval_loss": 0.7244272232055664,
|
| 216 |
+
"eval_runtime": 128.9211,
|
| 217 |
+
"eval_samples_per_second": 15.312,
|
| 218 |
+
"eval_steps_per_second": 1.916,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.8443093549476528,
|
| 223 |
+
"grad_norm": 0.30320531129837036,
|
| 224 |
+
"learning_rate": 2.155690645052347e-07,
|
| 225 |
+
"loss": 0.1551,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8443093549476528,
|
| 230 |
+
"eval_loss": 0.718368411064148,
|
| 231 |
+
"eval_runtime": 128.8149,
|
| 232 |
+
"eval_samples_per_second": 15.324,
|
| 233 |
+
"eval_steps_per_second": 1.917,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.9005966452774964,
|
| 238 |
+
"grad_norm": 1.9788849385804497e-05,
|
| 239 |
+
"learning_rate": 2.0994033547225037e-07,
|
| 240 |
+
"loss": 0.1439,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.9005966452774964,
|
| 245 |
+
"eval_loss": 0.7040167450904846,
|
| 246 |
+
"eval_runtime": 128.4921,
|
| 247 |
+
"eval_samples_per_second": 15.363,
|
| 248 |
+
"eval_steps_per_second": 1.922,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.9568839356073399,
|
| 253 |
+
"grad_norm": 1.0750063665909693e-06,
|
| 254 |
+
"learning_rate": 2.04311606439266e-07,
|
| 255 |
+
"loss": 0.1361,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.9568839356073399,
|
| 260 |
+
"eval_loss": 0.6983720660209656,
|
| 261 |
+
"eval_runtime": 128.5929,
|
| 262 |
+
"eval_samples_per_second": 15.351,
|
| 263 |
+
"eval_steps_per_second": 1.921,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.0131712259371835,
|
| 268 |
+
"grad_norm": 187.1415557861328,
|
| 269 |
+
"learning_rate": 1.9868287740628167e-07,
|
| 270 |
+
"loss": 0.1144,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.0131712259371835,
|
| 275 |
+
"eval_loss": 0.6901102066040039,
|
| 276 |
+
"eval_runtime": 128.5322,
|
| 277 |
+
"eval_samples_per_second": 15.358,
|
| 278 |
+
"eval_steps_per_second": 1.922,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.069458516267027,
|
| 283 |
+
"grad_norm": 0.007205183617770672,
|
| 284 |
+
"learning_rate": 1.9305414837329728e-07,
|
| 285 |
+
"loss": 0.0643,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.069458516267027,
|
| 290 |
+
"eval_loss": 0.6831753253936768,
|
| 291 |
+
"eval_runtime": 128.59,
|
| 292 |
+
"eval_samples_per_second": 15.351,
|
| 293 |
+
"eval_steps_per_second": 1.921,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.1257458065968704,
|
| 298 |
+
"grad_norm": 0.4961595833301544,
|
| 299 |
+
"learning_rate": 1.8742541934031294e-07,
|
| 300 |
+
"loss": 0.0773,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.1257458065968704,
|
| 305 |
+
"eval_loss": 0.6732445955276489,
|
| 306 |
+
"eval_runtime": 128.732,
|
| 307 |
+
"eval_samples_per_second": 15.334,
|
| 308 |
+
"eval_steps_per_second": 1.919,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.1820330969267139,
|
| 313 |
+
"grad_norm": 0.0009373470675200224,
|
| 314 |
+
"learning_rate": 1.817966903073286e-07,
|
| 315 |
+
"loss": 0.0814,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.1820330969267139,
|
| 320 |
+
"eval_loss": 0.6723716259002686,
|
| 321 |
+
"eval_runtime": 126.1133,
|
| 322 |
+
"eval_samples_per_second": 15.653,
|
| 323 |
+
"eval_steps_per_second": 1.959,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.2383203872565574,
|
| 328 |
+
"grad_norm": 2.7500348096509697e-06,
|
| 329 |
+
"learning_rate": 1.7616796127434424e-07,
|
| 330 |
+
"loss": 0.0924,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.2383203872565574,
|
| 335 |
+
"eval_loss": 0.6679245233535767,
|
| 336 |
+
"eval_runtime": 126.0452,
|
| 337 |
+
"eval_samples_per_second": 15.661,
|
| 338 |
+
"eval_steps_per_second": 1.96,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.294607677586401,
|
| 343 |
+
"grad_norm": 12.024052619934082,
|
| 344 |
+
"learning_rate": 1.705392322413599e-07,
|
| 345 |
+
"loss": 0.091,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.294607677586401,
|
| 350 |
+
"eval_loss": 0.6659140586853027,
|
| 351 |
+
"eval_runtime": 129.0611,
|
| 352 |
+
"eval_samples_per_second": 15.295,
|
| 353 |
+
"eval_steps_per_second": 1.914,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.3508949679162445,
|
| 358 |
+
"grad_norm": 0.0008855258929543197,
|
| 359 |
+
"learning_rate": 1.6491050320837554e-07,
|
| 360 |
+
"loss": 0.0978,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.3508949679162445,
|
| 365 |
+
"eval_loss": 0.6577230095863342,
|
| 366 |
+
"eval_runtime": 126.2858,
|
| 367 |
+
"eval_samples_per_second": 15.631,
|
| 368 |
+
"eval_steps_per_second": 1.956,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.407182258246088,
|
| 373 |
+
"grad_norm": 0.0003772446943912655,
|
| 374 |
+
"learning_rate": 1.592817741753912e-07,
|
| 375 |
+
"loss": 0.0598,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.407182258246088,
|
| 380 |
+
"eval_loss": 0.6527238488197327,
|
| 381 |
+
"eval_runtime": 125.8972,
|
| 382 |
+
"eval_samples_per_second": 15.679,
|
| 383 |
+
"eval_steps_per_second": 1.962,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.4634695485759315,
|
| 388 |
+
"grad_norm": 0.00032130314502865076,
|
| 389 |
+
"learning_rate": 1.5365304514240682e-07,
|
| 390 |
+
"loss": 0.0713,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.4634695485759315,
|
| 395 |
+
"eval_loss": 0.6472681760787964,
|
| 396 |
+
"eval_runtime": 128.5052,
|
| 397 |
+
"eval_samples_per_second": 15.361,
|
| 398 |
+
"eval_steps_per_second": 1.922,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.5197568389057752,
|
| 403 |
+
"grad_norm": 0.0006748048472218215,
|
| 404 |
+
"learning_rate": 1.4802431610942248e-07,
|
| 405 |
+
"loss": 0.0759,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.5197568389057752,
|
| 410 |
+
"eval_loss": 0.637607991695404,
|
| 411 |
+
"eval_runtime": 125.8757,
|
| 412 |
+
"eval_samples_per_second": 15.682,
|
| 413 |
+
"eval_steps_per_second": 1.962,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.5760441292356187,
|
| 418 |
+
"grad_norm": 4.276073184428242e-08,
|
| 419 |
+
"learning_rate": 1.4239558707643812e-07,
|
| 420 |
+
"loss": 0.093,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.5760441292356187,
|
| 425 |
+
"eval_loss": 0.6378567218780518,
|
| 426 |
+
"eval_runtime": 126.214,
|
| 427 |
+
"eval_samples_per_second": 15.64,
|
| 428 |
+
"eval_steps_per_second": 1.957,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.6323314195654621,
|
| 433 |
+
"grad_norm": 0.0019163701217621565,
|
| 434 |
+
"learning_rate": 1.3676685804345378e-07,
|
| 435 |
+
"loss": 0.061,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.6323314195654621,
|
| 440 |
+
"eval_loss": 0.6364826560020447,
|
| 441 |
+
"eval_runtime": 128.5083,
|
| 442 |
+
"eval_samples_per_second": 15.361,
|
| 443 |
+
"eval_steps_per_second": 1.922,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.6886187098953056,
|
| 448 |
+
"grad_norm": 10.94436264038086,
|
| 449 |
+
"learning_rate": 1.3113812901046944e-07,
|
| 450 |
+
"loss": 0.077,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.6886187098953056,
|
| 455 |
+
"eval_loss": 0.6351856589317322,
|
| 456 |
+
"eval_runtime": 126.3016,
|
| 457 |
+
"eval_samples_per_second": 15.629,
|
| 458 |
+
"eval_steps_per_second": 1.956,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.744906000225149,
|
| 463 |
+
"grad_norm": 245.41305541992188,
|
| 464 |
+
"learning_rate": 1.2550939997748508e-07,
|
| 465 |
+
"loss": 0.0798,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.744906000225149,
|
| 470 |
+
"eval_loss": 0.6253550052642822,
|
| 471 |
+
"eval_runtime": 126.2592,
|
| 472 |
+
"eval_samples_per_second": 15.635,
|
| 473 |
+
"eval_steps_per_second": 1.956,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.8011932905549926,
|
| 478 |
+
"grad_norm": 5.819400783479978e-08,
|
| 479 |
+
"learning_rate": 1.1988067094450072e-07,
|
| 480 |
+
"loss": 0.0846,
|
| 481 |
+
"step": 16000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.8011932905549926,
|
| 485 |
+
"eval_loss": 0.6173177361488342,
|
| 486 |
+
"eval_runtime": 128.7312,
|
| 487 |
+
"eval_samples_per_second": 15.334,
|
| 488 |
+
"eval_steps_per_second": 1.919,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.8574805808848363,
|
| 493 |
+
"grad_norm": 4.076898676430574e-06,
|
| 494 |
+
"learning_rate": 1.1425194191151638e-07,
|
| 495 |
+
"loss": 0.1003,
|
| 496 |
+
"step": 16500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.8574805808848363,
|
| 500 |
+
"eval_loss": 0.6105911135673523,
|
| 501 |
+
"eval_runtime": 126.0429,
|
| 502 |
+
"eval_samples_per_second": 15.661,
|
| 503 |
+
"eval_steps_per_second": 1.96,
|
| 504 |
+
"step": 16500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.9137678712146797,
|
| 508 |
+
"grad_norm": 2.8583364486694336,
|
| 509 |
+
"learning_rate": 1.0862321287853203e-07,
|
| 510 |
+
"loss": 0.0874,
|
| 511 |
+
"step": 17000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.9137678712146797,
|
| 515 |
+
"eval_loss": 0.6095408797264099,
|
| 516 |
+
"eval_runtime": 125.7222,
|
| 517 |
+
"eval_samples_per_second": 15.701,
|
| 518 |
+
"eval_steps_per_second": 1.965,
|
| 519 |
+
"step": 17000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.9700551615445232,
|
| 523 |
+
"grad_norm": 0.047957953065633774,
|
| 524 |
+
"learning_rate": 1.0299448384554767e-07,
|
| 525 |
+
"loss": 0.0513,
|
| 526 |
+
"step": 17500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.9700551615445232,
|
| 530 |
+
"eval_loss": 0.6062661409378052,
|
| 531 |
+
"eval_runtime": 128.3312,
|
| 532 |
+
"eval_samples_per_second": 15.382,
|
| 533 |
+
"eval_steps_per_second": 1.925,
|
| 534 |
+
"step": 17500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.026342451874367,
|
| 538 |
+
"grad_norm": 1.2455217301976518e-06,
|
| 539 |
+
"learning_rate": 9.736575481256332e-08,
|
| 540 |
+
"loss": 0.0743,
|
| 541 |
+
"step": 18000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.026342451874367,
|
| 545 |
+
"eval_loss": 0.6002796292304993,
|
| 546 |
+
"eval_runtime": 126.2895,
|
| 547 |
+
"eval_samples_per_second": 15.631,
|
| 548 |
+
"eval_steps_per_second": 1.956,
|
| 549 |
+
"step": 18000
|
| 550 |
+
}
|
| 551 |
+
],
|
| 552 |
+
"logging_steps": 500,
|
| 553 |
+
"max_steps": 26649,
|
| 554 |
+
"num_input_tokens_seen": 0,
|
| 555 |
+
"num_train_epochs": 3,
|
| 556 |
+
"save_steps": 500,
|
| 557 |
+
"stateful_callbacks": {
|
| 558 |
+
"TrainerControl": {
|
| 559 |
+
"args": {
|
| 560 |
+
"should_epoch_stop": false,
|
| 561 |
+
"should_evaluate": false,
|
| 562 |
+
"should_log": false,
|
| 563 |
+
"should_save": true,
|
| 564 |
+
"should_training_stop": false
|
| 565 |
+
},
|
| 566 |
+
"attributes": {}
|
| 567 |
+
}
|
| 568 |
+
},
|
| 569 |
+
"total_flos": 6476010081031260.0,
|
| 570 |
+
"train_batch_size": 2,
|
| 571 |
+
"trial_name": null,
|
| 572 |
+
"trial_params": null
|
| 573 |
+
}
|
checkpoint-18000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:074eb2b3aae1f3310fb0b27d83d259baa7ad74a32c8433c81af819e28558f6b0
|
| 3 |
+
size 4847
|