Training in progress, epoch 1

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<sep_secable>": 32000
+}

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 8.0,
+    "eval_loss": 0.025699105113744736,
+    "eval_mae": 0.110326386988163,
+    "eval_model_preparation_time": 0.0006,
+    "eval_mse": 0.02581467106938362,
+    "eval_runtime": 3.2416,
+    "eval_samples_per_second": 266.224,
+    "eval_steps_per_second": 33.317,
+    "total_flos": 145178794745856.0,
+    "train_loss": 0.046594451285070844,
+    "train_runtime": 251.4219,
+    "train_samples_per_second": 109.744,
+    "train_steps_per_second": 3.436
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 8.0,
+    "eval_loss": 0.025699105113744736,
+    "eval_mae": 0.110326386988163,
+    "eval_model_preparation_time": 0.0006,
+    "eval_mse": 0.02581467106938362,
+    "eval_runtime": 3.2416,
+    "eval_samples_per_second": 266.224,
+    "eval_steps_per_second": 33.317
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94c27b4f84078381cad2539584a911d3a0ce319f8bf5a6b25b01b2579bb33abb
+oid sha256:ef04824a65d04c957646b3858dacd194db24dd852033c401441b339a4b111bfa
 size 23502148

tokenizer.json

@@ -47,6 +47,15 @@
       "rstrip": false,
       "normalized": false,
       "special": true
+    },
+    {
+      "id": 32000,
+      "content": "<sep_secable>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
     }
   ],
   "normalizer": null,

tokenizer_config.json

@@ -40,6 +40,14 @@
       "rstrip": false,
       "single_word": false,
       "special": true
+    },
+    "32000": {
+      "content": "<sep_secable>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
     }
   },
   "bos_token": "<s>",

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "total_flos": 145178794745856.0,
+    "train_loss": 0.046594451285070844,
+    "train_runtime": 251.4219,
+    "train_samples_per_second": 109.744,
+    "train_steps_per_second": 3.436
+}

trainer_state.json

@@ -0,0 +1,250 @@
+{
+  "best_global_step": 864,
+  "best_metric": 0.02581467106938362,
+  "best_model_checkpoint": "./bert-secabilite-regressor/checkpoint-864",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 864,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.46296296296296297,
+      "grad_norm": 0.18632853031158447,
+      "learning_rate": 2.8298611111111113e-05,
+      "loss": 0.1324,
+      "step": 50
+    },
+    {
+      "epoch": 0.9259259259259259,
+      "grad_norm": 0.16045570373535156,
+      "learning_rate": 2.65625e-05,
+      "loss": 0.0928,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.0562266930937767,
+      "eval_mae": 0.19059279561042786,
+      "eval_model_preparation_time": 0.0006,
+      "eval_mse": 0.056342288851737976,
+      "eval_runtime": 3.1074,
+      "eval_samples_per_second": 277.724,
+      "eval_steps_per_second": 34.756,
+      "step": 108
+    },
+    {
+      "epoch": 1.3888888888888888,
+      "grad_norm": 0.14682810008525848,
+      "learning_rate": 2.4826388888888888e-05,
+      "loss": 0.0641,
+      "step": 150
+    },
+    {
+      "epoch": 1.8518518518518519,
+      "grad_norm": 0.1716427206993103,
+      "learning_rate": 2.3090277777777777e-05,
+      "loss": 0.0531,
+      "step": 200
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.03785160928964615,
+      "eval_mae": 0.1461946666240692,
+      "eval_model_preparation_time": 0.0006,
+      "eval_mse": 0.03801807761192322,
+      "eval_runtime": 2.9858,
+      "eval_samples_per_second": 289.032,
+      "eval_steps_per_second": 36.171,
+      "step": 216
+    },
+    {
+      "epoch": 2.314814814814815,
+      "grad_norm": 0.13463342189788818,
+      "learning_rate": 2.135416666666667e-05,
+      "loss": 0.0465,
+      "step": 250
+    },
+    {
+      "epoch": 2.7777777777777777,
+      "grad_norm": 0.12161467969417572,
+      "learning_rate": 1.961805555555556e-05,
+      "loss": 0.0427,
+      "step": 300
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.03144295513629913,
+      "eval_mae": 0.1291944533586502,
+      "eval_model_preparation_time": 0.0006,
+      "eval_mse": 0.031581759452819824,
+      "eval_runtime": 2.9834,
+      "eval_samples_per_second": 289.266,
+      "eval_steps_per_second": 36.2,
+      "step": 324
+    },
+    {
+      "epoch": 3.240740740740741,
+      "grad_norm": 0.12704914808273315,
+      "learning_rate": 1.7881944444444445e-05,
+      "loss": 0.0387,
+      "step": 350
+    },
+    {
+      "epoch": 3.7037037037037037,
+      "grad_norm": 0.09777373820543289,
+      "learning_rate": 1.6145833333333334e-05,
+      "loss": 0.037,
+      "step": 400
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.028512364253401756,
+      "eval_mae": 0.12053852528333664,
+      "eval_model_preparation_time": 0.0006,
+      "eval_mse": 0.02863851748406887,
+      "eval_runtime": 2.8921,
+      "eval_samples_per_second": 298.403,
+      "eval_steps_per_second": 37.344,
+      "step": 432
+    },
+    {
+      "epoch": 4.166666666666667,
+      "grad_norm": 0.11912214756011963,
+      "learning_rate": 1.4409722222222223e-05,
+      "loss": 0.035,
+      "step": 450
+    },
+    {
+      "epoch": 4.62962962962963,
+      "grad_norm": 0.09360779076814651,
+      "learning_rate": 1.2673611111111112e-05,
+      "loss": 0.0342,
+      "step": 500
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.027181362733244896,
+      "eval_mae": 0.11569906771183014,
+      "eval_model_preparation_time": 0.0006,
+      "eval_mse": 0.027303453534841537,
+      "eval_runtime": 2.9849,
+      "eval_samples_per_second": 289.119,
+      "eval_steps_per_second": 36.182,
+      "step": 540
+    },
+    {
+      "epoch": 5.092592592592593,
+      "grad_norm": 0.10609101504087448,
+      "learning_rate": 1.09375e-05,
+      "loss": 0.0327,
+      "step": 550
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 0.10040564835071564,
+      "learning_rate": 9.201388888888889e-06,
+      "loss": 0.0321,
+      "step": 600
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.026255721226334572,
+      "eval_mae": 0.11266371607780457,
+      "eval_model_preparation_time": 0.0006,
+      "eval_mse": 0.026374226436018944,
+      "eval_runtime": 3.0505,
+      "eval_samples_per_second": 282.907,
+      "eval_steps_per_second": 35.404,
+      "step": 648
+    },
+    {
+      "epoch": 6.018518518518518,
+      "grad_norm": 0.08536071330308914,
+      "learning_rate": 7.4652777777777775e-06,
+      "loss": 0.0321,
+      "step": 650
+    },
+    {
+      "epoch": 6.481481481481482,
+      "grad_norm": 0.10104560106992722,
+      "learning_rate": 5.729166666666667e-06,
+      "loss": 0.031,
+      "step": 700
+    },
+    {
+      "epoch": 6.944444444444445,
+      "grad_norm": 0.1454145610332489,
+      "learning_rate": 3.993055555555556e-06,
+      "loss": 0.0311,
+      "step": 750
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.025880420580506325,
+      "eval_mae": 0.11090163141489029,
+      "eval_model_preparation_time": 0.0006,
+      "eval_mse": 0.025996480137109756,
+      "eval_runtime": 2.9444,
+      "eval_samples_per_second": 293.101,
+      "eval_steps_per_second": 36.68,
+      "step": 756
+    },
+    {
+      "epoch": 7.407407407407407,
+      "grad_norm": 0.10250015556812286,
+      "learning_rate": 2.2569444444444443e-06,
+      "loss": 0.0304,
+      "step": 800
+    },
+    {
+      "epoch": 7.87037037037037,
+      "grad_norm": 0.08759289234876633,
+      "learning_rate": 5.208333333333334e-07,
+      "loss": 0.0306,
+      "step": 850
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.025699105113744736,
+      "eval_mae": 0.110326386988163,
+      "eval_model_preparation_time": 0.0006,
+      "eval_mse": 0.02581467106938362,
+      "eval_runtime": 2.9499,
+      "eval_samples_per_second": 292.55,
+      "eval_steps_per_second": 36.611,
+      "step": 864
+    },
+    {
+      "epoch": 8.0,
+      "step": 864,
+      "total_flos": 145178794745856.0,
+      "train_loss": 0.046594451285070844,
+      "train_runtime": 251.4219,
+      "train_samples_per_second": 109.744,
+      "train_steps_per_second": 3.436
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 864,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 145178794745856.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}