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 sha256:3cb1fe8307ec6f22e015ca4870cbdfb1c7eec62d6e0757fbe2e928472f0c8057
|
| 3 |
+
size 1711848436
|
checkpoint-14500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42d23c39e59cff80319678261df759c667a3b9bb78404de3e1c13654adc4e857
|
| 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:e8681bf33a0cfabf27b1c43af8f022a52f0079701b912daa386aebb8f9118515
|
| 3 |
+
size 14503
|
checkpoint-14500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e0faad27c3d302805f2cc02025cbeb0d37fe190c4fa396d39d22a0d28557297
|
| 3 |
+
size 623
|
checkpoint-14500/trainer_state.json
ADDED
|
@@ -0,0 +1,468 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9193492531776428,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e6l56-l/checkpoint-12500",
|
| 4 |
+
"epoch": 2.671333824613117,
|
| 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": 0.09211495946941783,
|
| 13 |
+
"grad_norm": 0.49671515822410583,
|
| 14 |
+
"learning_rate": 4.923237533775486e-06,
|
| 15 |
+
"loss": 0.2159,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09211495946941783,
|
| 20 |
+
"eval_loss": 1.4818799495697021,
|
| 21 |
+
"eval_runtime": 72.2844,
|
| 22 |
+
"eval_samples_per_second": 16.698,
|
| 23 |
+
"eval_steps_per_second": 2.089,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.18422991893883567,
|
| 28 |
+
"grad_norm": 0.0015633106231689453,
|
| 29 |
+
"learning_rate": 4.846475067550971e-06,
|
| 30 |
+
"loss": 0.2648,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.18422991893883567,
|
| 35 |
+
"eval_loss": 1.4225908517837524,
|
| 36 |
+
"eval_runtime": 73.6999,
|
| 37 |
+
"eval_samples_per_second": 16.377,
|
| 38 |
+
"eval_steps_per_second": 2.049,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.2763448784082535,
|
| 43 |
+
"grad_norm": 0.047471947968006134,
|
| 44 |
+
"learning_rate": 4.769712601326456e-06,
|
| 45 |
+
"loss": 0.328,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.2763448784082535,
|
| 50 |
+
"eval_loss": 1.412400245666504,
|
| 51 |
+
"eval_runtime": 74.9915,
|
| 52 |
+
"eval_samples_per_second": 16.095,
|
| 53 |
+
"eval_steps_per_second": 2.014,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.36845983787767134,
|
| 58 |
+
"grad_norm": 3.9284778208248383e-10,
|
| 59 |
+
"learning_rate": 4.692950135101941e-06,
|
| 60 |
+
"loss": 0.1813,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.36845983787767134,
|
| 65 |
+
"eval_loss": 1.2844351530075073,
|
| 66 |
+
"eval_runtime": 74.7649,
|
| 67 |
+
"eval_samples_per_second": 16.144,
|
| 68 |
+
"eval_steps_per_second": 2.02,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.46057479734708917,
|
| 73 |
+
"grad_norm": 514.6055297851562,
|
| 74 |
+
"learning_rate": 4.616187668877426e-06,
|
| 75 |
+
"loss": 0.4836,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.46057479734708917,
|
| 80 |
+
"eval_loss": 1.301351547241211,
|
| 81 |
+
"eval_runtime": 74.7994,
|
| 82 |
+
"eval_samples_per_second": 16.136,
|
| 83 |
+
"eval_steps_per_second": 2.019,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.552689756816507,
|
| 88 |
+
"grad_norm": 127.98405456542969,
|
| 89 |
+
"learning_rate": 4.5394252026529115e-06,
|
| 90 |
+
"loss": 0.2803,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.552689756816507,
|
| 95 |
+
"eval_loss": 1.1920161247253418,
|
| 96 |
+
"eval_runtime": 75.5447,
|
| 97 |
+
"eval_samples_per_second": 15.977,
|
| 98 |
+
"eval_steps_per_second": 1.999,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.6448047162859248,
|
| 103 |
+
"grad_norm": 0.048586923629045486,
|
| 104 |
+
"learning_rate": 4.462662736428396e-06,
|
| 105 |
+
"loss": 0.2056,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.6448047162859248,
|
| 110 |
+
"eval_loss": 1.1428992748260498,
|
| 111 |
+
"eval_runtime": 75.7202,
|
| 112 |
+
"eval_samples_per_second": 15.94,
|
| 113 |
+
"eval_steps_per_second": 1.994,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7369196757553427,
|
| 118 |
+
"grad_norm": 228.63064575195312,
|
| 119 |
+
"learning_rate": 4.385900270203882e-06,
|
| 120 |
+
"loss": 0.2174,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7369196757553427,
|
| 125 |
+
"eval_loss": 1.1957736015319824,
|
| 126 |
+
"eval_runtime": 74.1933,
|
| 127 |
+
"eval_samples_per_second": 16.268,
|
| 128 |
+
"eval_steps_per_second": 2.035,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.8290346352247605,
|
| 133 |
+
"grad_norm": 226.2799530029297,
|
| 134 |
+
"learning_rate": 4.3091378039793666e-06,
|
| 135 |
+
"loss": 0.2236,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.8290346352247605,
|
| 140 |
+
"eval_loss": 1.0961644649505615,
|
| 141 |
+
"eval_runtime": 74.6968,
|
| 142 |
+
"eval_samples_per_second": 16.159,
|
| 143 |
+
"eval_steps_per_second": 2.022,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.9211495946941783,
|
| 148 |
+
"grad_norm": 8.084862640478363e-12,
|
| 149 |
+
"learning_rate": 4.232375337754851e-06,
|
| 150 |
+
"loss": 0.1966,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.9211495946941783,
|
| 155 |
+
"eval_loss": 1.1026136875152588,
|
| 156 |
+
"eval_runtime": 75.566,
|
| 157 |
+
"eval_samples_per_second": 15.973,
|
| 158 |
+
"eval_steps_per_second": 1.998,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.013264554163596,
|
| 163 |
+
"grad_norm": 1.592839794284373e-06,
|
| 164 |
+
"learning_rate": 4.155612871530337e-06,
|
| 165 |
+
"loss": 0.2547,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.013264554163596,
|
| 170 |
+
"eval_loss": 1.0125067234039307,
|
| 171 |
+
"eval_runtime": 74.4786,
|
| 172 |
+
"eval_samples_per_second": 16.206,
|
| 173 |
+
"eval_steps_per_second": 2.027,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.105379513633014,
|
| 178 |
+
"grad_norm": 0.06377054750919342,
|
| 179 |
+
"learning_rate": 4.078850405305822e-06,
|
| 180 |
+
"loss": 0.1057,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.105379513633014,
|
| 185 |
+
"eval_loss": 1.1245251893997192,
|
| 186 |
+
"eval_runtime": 74.102,
|
| 187 |
+
"eval_samples_per_second": 16.288,
|
| 188 |
+
"eval_steps_per_second": 2.038,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.1974944731024317,
|
| 193 |
+
"grad_norm": 0.00903751328587532,
|
| 194 |
+
"learning_rate": 4.002087939081307e-06,
|
| 195 |
+
"loss": 0.1085,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.1974944731024317,
|
| 200 |
+
"eval_loss": 1.1834831237792969,
|
| 201 |
+
"eval_runtime": 74.4532,
|
| 202 |
+
"eval_samples_per_second": 16.212,
|
| 203 |
+
"eval_steps_per_second": 2.028,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.2896094325718497,
|
| 208 |
+
"grad_norm": 5.0311258631419165e-11,
|
| 209 |
+
"learning_rate": 3.925325472856792e-06,
|
| 210 |
+
"loss": 0.1174,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.2896094325718497,
|
| 215 |
+
"eval_loss": 1.1334316730499268,
|
| 216 |
+
"eval_runtime": 74.1756,
|
| 217 |
+
"eval_samples_per_second": 16.272,
|
| 218 |
+
"eval_steps_per_second": 2.036,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.3817243920412676,
|
| 223 |
+
"grad_norm": 3.813414917175351e-09,
|
| 224 |
+
"learning_rate": 3.848563006632278e-06,
|
| 225 |
+
"loss": 0.122,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.3817243920412676,
|
| 230 |
+
"eval_loss": 1.1058012247085571,
|
| 231 |
+
"eval_runtime": 74.0285,
|
| 232 |
+
"eval_samples_per_second": 16.305,
|
| 233 |
+
"eval_steps_per_second": 2.04,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.4738393515106853,
|
| 238 |
+
"grad_norm": 4.6230165935412515e-06,
|
| 239 |
+
"learning_rate": 3.7718005404077624e-06,
|
| 240 |
+
"loss": 0.1653,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.4738393515106853,
|
| 245 |
+
"eval_loss": 1.8118412494659424,
|
| 246 |
+
"eval_runtime": 73.9465,
|
| 247 |
+
"eval_samples_per_second": 16.323,
|
| 248 |
+
"eval_steps_per_second": 2.042,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.565954310980103,
|
| 253 |
+
"grad_norm": 6.103370031596e-10,
|
| 254 |
+
"learning_rate": 3.6950380741832476e-06,
|
| 255 |
+
"loss": 0.1144,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.565954310980103,
|
| 260 |
+
"eval_loss": 1.0845656394958496,
|
| 261 |
+
"eval_runtime": 73.8897,
|
| 262 |
+
"eval_samples_per_second": 16.335,
|
| 263 |
+
"eval_steps_per_second": 2.044,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.658069270449521,
|
| 268 |
+
"grad_norm": 0.023263949900865555,
|
| 269 |
+
"learning_rate": 3.6182756079587327e-06,
|
| 270 |
+
"loss": 0.1866,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.658069270449521,
|
| 275 |
+
"eval_loss": 1.2360613346099854,
|
| 276 |
+
"eval_runtime": 73.8786,
|
| 277 |
+
"eval_samples_per_second": 16.338,
|
| 278 |
+
"eval_steps_per_second": 2.044,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.750184229918939,
|
| 283 |
+
"grad_norm": 1.1146563672070897e-09,
|
| 284 |
+
"learning_rate": 3.541513141734218e-06,
|
| 285 |
+
"loss": 0.1385,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.750184229918939,
|
| 290 |
+
"eval_loss": 1.0680217742919922,
|
| 291 |
+
"eval_runtime": 73.8064,
|
| 292 |
+
"eval_samples_per_second": 16.354,
|
| 293 |
+
"eval_steps_per_second": 2.046,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.8422991893883567,
|
| 298 |
+
"grad_norm": 0.0801646038889885,
|
| 299 |
+
"learning_rate": 3.464750675509703e-06,
|
| 300 |
+
"loss": 0.1178,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.8422991893883567,
|
| 305 |
+
"eval_loss": 0.9592632055282593,
|
| 306 |
+
"eval_runtime": 73.9232,
|
| 307 |
+
"eval_samples_per_second": 16.328,
|
| 308 |
+
"eval_steps_per_second": 2.043,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.9344141488577744,
|
| 313 |
+
"grad_norm": 8.68252456598384e-08,
|
| 314 |
+
"learning_rate": 3.3879882092851883e-06,
|
| 315 |
+
"loss": 0.1506,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.9344141488577744,
|
| 320 |
+
"eval_loss": 1.0038937330245972,
|
| 321 |
+
"eval_runtime": 74.3344,
|
| 322 |
+
"eval_samples_per_second": 16.237,
|
| 323 |
+
"eval_steps_per_second": 2.031,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.026529108327192,
|
| 328 |
+
"grad_norm": 1.1008462763584248e-07,
|
| 329 |
+
"learning_rate": 3.311225743060673e-06,
|
| 330 |
+
"loss": 0.0938,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.026529108327192,
|
| 335 |
+
"eval_loss": 0.931797206401825,
|
| 336 |
+
"eval_runtime": 74.2035,
|
| 337 |
+
"eval_samples_per_second": 16.266,
|
| 338 |
+
"eval_steps_per_second": 2.035,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.1186440677966103,
|
| 343 |
+
"grad_norm": 7.901914614194538e-06,
|
| 344 |
+
"learning_rate": 3.234463276836158e-06,
|
| 345 |
+
"loss": 0.093,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.1186440677966103,
|
| 350 |
+
"eval_loss": 0.9992227554321289,
|
| 351 |
+
"eval_runtime": 74.2516,
|
| 352 |
+
"eval_samples_per_second": 16.256,
|
| 353 |
+
"eval_steps_per_second": 2.034,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.210759027266028,
|
| 358 |
+
"grad_norm": 0.04380282014608383,
|
| 359 |
+
"learning_rate": 3.1577008106116434e-06,
|
| 360 |
+
"loss": 0.0877,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.210759027266028,
|
| 365 |
+
"eval_loss": 0.9989096522331238,
|
| 366 |
+
"eval_runtime": 75.2761,
|
| 367 |
+
"eval_samples_per_second": 16.034,
|
| 368 |
+
"eval_steps_per_second": 2.006,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.3028739867354457,
|
| 373 |
+
"grad_norm": 65.85383605957031,
|
| 374 |
+
"learning_rate": 3.080938344387129e-06,
|
| 375 |
+
"loss": 0.1102,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 2.3028739867354457,
|
| 380 |
+
"eval_loss": 0.9193492531776428,
|
| 381 |
+
"eval_runtime": 74.3653,
|
| 382 |
+
"eval_samples_per_second": 16.231,
|
| 383 |
+
"eval_steps_per_second": 2.031,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.3949889462048635,
|
| 388 |
+
"grad_norm": 0.0006995272124186158,
|
| 389 |
+
"learning_rate": 3.004175878162614e-06,
|
| 390 |
+
"loss": 0.0916,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.3949889462048635,
|
| 395 |
+
"eval_loss": 0.9769579172134399,
|
| 396 |
+
"eval_runtime": 74.2536,
|
| 397 |
+
"eval_samples_per_second": 16.255,
|
| 398 |
+
"eval_steps_per_second": 2.034,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.4871039056742816,
|
| 403 |
+
"grad_norm": 3.747671507881023e-05,
|
| 404 |
+
"learning_rate": 2.9274134119380993e-06,
|
| 405 |
+
"loss": 0.0665,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.4871039056742816,
|
| 410 |
+
"eval_loss": 0.982865035533905,
|
| 411 |
+
"eval_runtime": 75.5772,
|
| 412 |
+
"eval_samples_per_second": 15.97,
|
| 413 |
+
"eval_steps_per_second": 1.998,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.5792188651436994,
|
| 418 |
+
"grad_norm": 4.0381178855896,
|
| 419 |
+
"learning_rate": 2.850650945713584e-06,
|
| 420 |
+
"loss": 0.0545,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.5792188651436994,
|
| 425 |
+
"eval_loss": 0.9746910929679871,
|
| 426 |
+
"eval_runtime": 74.7109,
|
| 427 |
+
"eval_samples_per_second": 16.156,
|
| 428 |
+
"eval_steps_per_second": 2.021,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.671333824613117,
|
| 433 |
+
"grad_norm": 7.164443260876396e-09,
|
| 434 |
+
"learning_rate": 2.7738884794890693e-06,
|
| 435 |
+
"loss": 0.1023,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.671333824613117,
|
| 440 |
+
"eval_loss": 0.9872701168060303,
|
| 441 |
+
"eval_runtime": 73.8157,
|
| 442 |
+
"eval_samples_per_second": 16.352,
|
| 443 |
+
"eval_steps_per_second": 2.046,
|
| 444 |
+
"step": 14500
|
| 445 |
+
}
|
| 446 |
+
],
|
| 447 |
+
"logging_steps": 500,
|
| 448 |
+
"max_steps": 32568,
|
| 449 |
+
"num_input_tokens_seen": 0,
|
| 450 |
+
"num_train_epochs": 6,
|
| 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 sha256:09987e134b23188d490a3f3d609c84b1bd02ce771791cf447dcb77366fe93878
|
| 3 |
+
size 4847
|