Next training iteration (2.33m)

added_tokens.json

@@ -1 +1 @@
-{"PROFANITY_TOKEN": 32109, "END_INTERACTION_TOKEN": 32118, "URL_TOKEN": 32100, "EXTRACT_SEGMENTS: ": 32110, "NUMBER_TOKEN": 32103, "NUMBER_PERCENTAGE_TOKEN": 32102, "LONG_WORD_TOKEN": 32105, "END_SPONSOR_TOKEN": 32114, "START_SELFPROMO_TOKEN": 32115, "END_SELFPROMO_TOKEN": 32116, "NO_SEGMENT_TOKEN": 32112, "HYPHENATED_URL_TOKEN": 32101, "START_SPONSOR_TOKEN": 32113, "[Music]": 32106, "[Laughter]": 32108, "BETWEEN_SEGMENTS_TOKEN": 32111, "[Applause]": 32107, "START_INTERACTION_TOKEN": 32117, "SHORT_HYPHENATED_TOKEN": 32104}
+{"NUMBER_TOKEN": 32103, "EXTRACT_SEGMENTS: ": 32110, "URL_TOKEN": 32100, "START_SPONSOR_TOKEN": 32113, "START_SELFPROMO_TOKEN": 32115, "SHORT_HYPHENATED_TOKEN": 32104, "NO_SEGMENT_TOKEN": 32112, "START_INTERACTION_TOKEN": 32117, "[Music]": 32106, "[Applause]": 32107, "BETWEEN_SEGMENTS_TOKEN": 32111, "END_SPONSOR_TOKEN": 32114, "[Laughter]": 32108, "LONG_WORD_TOKEN": 32105, "END_SELFPROMO_TOKEN": 32116, "END_INTERACTION_TOKEN": 32118, "NUMBER_PERCENTAGE_TOKEN": 32102, "HYPHENATED_URL_TOKEN": 32101, "PROFANITY_TOKEN": 32109}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "models/small-1712500/",
+  "_name_or_path": "models/small+140000/",
   "architectures": [
     "T5ForConditionalGeneration"
   ],

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49525bef46229b9dd14d256aab3b299a9abc1fe2f92316ca7465eda817803842
+oid sha256:953853624a364416ef1c00d39cc455ef45fe4a965b17640ecd2e284e286ea198
 size 307892869

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6019e275337380401ff115e5b151ec62b90d64182c9e491d19f52523b075b522
+oid sha256:78a98b37c4fbbb2d3fe519fa6ad3f69d142d52742de56fcb0d1cd9eb549a0a04
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aeafd2e8b1e45dc5c6795cc90be9dfd1344b28195e3f803fd812afbd480b4467
+oid sha256:7a7df15d137fce618f6fe590ee2b8930bb7a89e41510e95f6ea420bc96e1b47d
 size 623

tokenizer_config.json

@@ -1 +1 @@
-{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "model_max_length": 512, "name_or_path": "models/small-1712500/", "special_tokens_map_file": "C:\\Users\\joshu/.cache\\huggingface\\transformers\\3ad6f8335c1b1ef8966245899d47dcf735abd134d21fd7d26f621fe45ac01184.c94798918c92ded6aeef2d2f0e666d2cc4145eca1aa6e1336fde07f2e13e2f46", "sp_model_kwargs": {}, "max_length": 512, "tokenizer_class": "T5Tokenizer"}
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "model_max_length": 512, "name_or_path": "models/small+140000/", "special_tokens_map_file": "C:\\Users\\joshu/.cache\\huggingface\\transformers\\3ad6f8335c1b1ef8966245899d47dcf735abd134d21fd7d26f621fe45ac01184.c94798918c92ded6aeef2d2f0e666d2cc4145eca1aa6e1336fde07f2e13e2f46", "sp_model_kwargs": {}, "max_length": 512, "tokenizer_class": "T5Tokenizer"}

trainer_state.json

@@ -1,236 +1,306 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.678837487353996,
-  "global_step": 320000,
+  "epoch": 1.6139607605790083,
+  "global_step": 160000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.06,
-      "learning_rate": 4.971259082130047e-05,
-      "loss": 0.0475,
+      "epoch": 0.05,
+      "learning_rate": 4.7478186311595304e-05,
+      "loss": 0.0451,
       "step": 5000
     },
     {
-      "epoch": 0.11,
-      "learning_rate": 4.9425181642600946e-05,
-      "loss": 0.0494,
+      "epoch": 0.05,
+      "eval_loss": 0.04004961997270584,
+      "eval_runtime": 280.6429,
+      "eval_samples_per_second": 78.498,
+      "eval_steps_per_second": 19.626,
+      "step": 5000
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.49563726231906e-05,
+      "loss": 0.0479,
       "step": 10000
     },
     {
-      "epoch": 0.11,
-      "eval_loss": 0.04334285110235214,
-      "eval_runtime": 233.2791,
-      "eval_samples_per_second": 82.862,
-      "eval_steps_per_second": 20.718,
+      "epoch": 0.1,
+      "eval_loss": 0.040193621069192886,
+      "eval_runtime": 274.5073,
+      "eval_samples_per_second": 80.253,
+      "eval_steps_per_second": 20.065,
       "step": 10000
     },
     {
-      "epoch": 0.17,
-      "learning_rate": 4.913777246390141e-05,
-      "loss": 0.049,
+      "epoch": 0.15,
+      "learning_rate": 4.24345589347859e-05,
+      "loss": 0.0442,
       "step": 15000
     },
     {
-      "epoch": 0.23,
-      "learning_rate": 4.8850363285201876e-05,
-      "loss": 0.0477,
+      "epoch": 0.15,
+      "eval_loss": 0.040153153240680695,
+      "eval_runtime": 270.832,
+      "eval_samples_per_second": 81.342,
+      "eval_steps_per_second": 20.337,
+      "step": 15000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 3.99127452463812e-05,
+      "loss": 0.0443,
       "step": 20000
     },
     {
-      "epoch": 0.23,
-      "eval_loss": 0.04320817440748215,
-      "eval_runtime": 230.0333,
-      "eval_samples_per_second": 84.031,
-      "eval_steps_per_second": 21.01,
+      "epoch": 0.2,
+      "eval_loss": 0.04070935398340225,
+      "eval_runtime": 270.5632,
+      "eval_samples_per_second": 81.423,
+      "eval_steps_per_second": 20.358,
       "step": 20000
     },
     {
-      "epoch": 0.29,
-      "learning_rate": 4.856295410650235e-05,
-      "loss": 0.0494,
+      "epoch": 0.25,
+      "learning_rate": 3.73909315579765e-05,
+      "loss": 0.044,
       "step": 25000
     },
     {
-      "epoch": 0.34,
-      "learning_rate": 4.827554492780282e-05,
-      "loss": 0.0464,
+      "epoch": 0.25,
+      "eval_loss": 0.0404120609164238,
+      "eval_runtime": 270.4511,
+      "eval_samples_per_second": 81.456,
+      "eval_steps_per_second": 20.366,
+      "step": 25000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.48691178695718e-05,
+      "loss": 0.0469,
       "step": 30000
     },
     {
-      "epoch": 0.34,
-      "eval_loss": 0.04396720230579376,
-      "eval_runtime": 235.2064,
-      "eval_samples_per_second": 82.183,
-      "eval_steps_per_second": 20.548,
+      "epoch": 0.3,
+      "eval_loss": 0.03903116658329964,
+      "eval_runtime": 270.9455,
+      "eval_samples_per_second": 81.308,
+      "eval_steps_per_second": 20.329,
       "step": 30000
     },
     {
-      "epoch": 0.4,
-      "learning_rate": 4.7988135749103285e-05,
-      "loss": 0.0491,
+      "epoch": 0.35,
+      "learning_rate": 3.23473041811671e-05,
+      "loss": 0.0464,
       "step": 35000
     },
     {
-      "epoch": 0.46,
-      "learning_rate": 4.770072657040375e-05,
-      "loss": 0.0472,
+      "epoch": 0.35,
+      "eval_loss": 0.039902038872241974,
+      "eval_runtime": 271.2781,
+      "eval_samples_per_second": 81.208,
+      "eval_steps_per_second": 20.304,
+      "step": 35000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 2.9825490492762397e-05,
+      "loss": 0.0454,
       "step": 40000
     },
     {
-      "epoch": 0.46,
-      "eval_loss": 0.04489925131201744,
-      "eval_runtime": 225.0238,
-      "eval_samples_per_second": 85.902,
-      "eval_steps_per_second": 21.478,
+      "epoch": 0.4,
+      "eval_loss": 0.04037521034479141,
+      "eval_runtime": 270.9003,
+      "eval_samples_per_second": 81.321,
+      "eval_steps_per_second": 20.332,
       "step": 40000
     },
     {
-      "epoch": 0.69,
-      "learning_rate": 4.655108985560563e-05,
-      "loss": 0.0484,
-      "step": 60000
+      "epoch": 0.45,
+      "learning_rate": 2.7303676804357698e-05,
+      "loss": 0.0446,
+      "step": 45000
     },
     {
-      "epoch": 0.92,
-      "learning_rate": 4.540145314080751e-05,
-      "loss": 0.0481,
-      "step": 80000
+      "epoch": 0.45,
+      "eval_loss": 0.03930870071053505,
+      "eval_runtime": 274.5685,
+      "eval_samples_per_second": 80.235,
+      "eval_steps_per_second": 20.061,
+      "step": 45000
     },
     {
-      "epoch": 0.92,
-      "eval_loss": 0.044077709317207336,
-      "eval_runtime": 226.2289,
-      "eval_samples_per_second": 85.444,
-      "eval_steps_per_second": 21.363,
-      "step": 80000
+      "epoch": 0.5,
+      "learning_rate": 2.4781863115952993e-05,
+      "loss": 0.0426,
+      "step": 50000
     },
     {
-      "epoch": 1.15,
-      "learning_rate": 4.4251816426009386e-05,
-      "loss": 0.0483,
-      "step": 100000
+      "epoch": 0.5,
+      "eval_loss": 0.03955301269888878,
+      "eval_runtime": 267.684,
+      "eval_samples_per_second": 82.299,
+      "eval_steps_per_second": 20.576,
+      "step": 50000
     },
     {
-      "epoch": 1.38,
-      "learning_rate": 4.310217971121126e-05,
-      "loss": 0.0463,
-      "step": 120000
+      "epoch": 0.55,
+      "learning_rate": 2.2260049427548294e-05,
+      "loss": 0.0444,
+      "step": 55000
     },
     {
-      "epoch": 1.38,
-      "eval_loss": 0.04454144090414047,
-      "eval_runtime": 228.877,
-      "eval_samples_per_second": 84.456,
-      "eval_steps_per_second": 21.116,
-      "step": 120000
+      "epoch": 0.55,
+      "eval_loss": 0.03960525989532471,
+      "eval_runtime": 267.9101,
+      "eval_samples_per_second": 82.229,
+      "eval_steps_per_second": 20.559,
+      "step": 55000
     },
     {
-      "epoch": 1.61,
-      "learning_rate": 4.1952542996413133e-05,
-      "loss": 0.0467,
-      "step": 140000
+      "epoch": 0.61,
+      "learning_rate": 1.9738235739143592e-05,
+      "loss": 0.0436,
+      "step": 60000
     },
     {
-      "epoch": 1.84,
-      "learning_rate": 4.0802906281615014e-05,
-      "loss": 0.0465,
-      "step": 160000
+      "epoch": 0.61,
+      "eval_loss": 0.039675917476415634,
+      "eval_runtime": 267.1715,
+      "eval_samples_per_second": 82.456,
+      "eval_steps_per_second": 20.616,
+      "step": 60000
     },
     {
-      "epoch": 1.84,
-      "eval_loss": 0.04413667321205139,
-      "eval_runtime": 228.9059,
-      "eval_samples_per_second": 84.445,
-      "eval_steps_per_second": 21.113,
-      "step": 160000
+      "epoch": 0.66,
+      "learning_rate": 1.7216422050738893e-05,
+      "loss": 0.0438,
+      "step": 65000
     },
     {
-      "epoch": 2.07,
-      "learning_rate": 3.965326956681689e-05,
-      "loss": 0.0469,
-      "step": 180000
+      "epoch": 0.66,
+      "eval_loss": 0.03922617807984352,
+      "eval_runtime": 275.6861,
+      "eval_samples_per_second": 79.91,
+      "eval_steps_per_second": 19.979,
+      "step": 65000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4694608362334191e-05,
+      "loss": 0.0429,
+      "step": 70000
+    },
+    {
+      "epoch": 0.71,
+      "eval_loss": 0.03912888839840889,
+      "eval_runtime": 268.88,
+      "eval_samples_per_second": 81.932,
+      "eval_steps_per_second": 20.485,
+      "step": 70000
     },
     {
-      "epoch": 2.3,
-      "learning_rate": 3.850363285201876e-05,
-      "loss": 0.0447,
-      "step": 200000
+      "epoch": 0.81,
+      "learning_rate": 4.5965098098552486e-05,
+      "loss": 0.044,
+      "step": 80000
     },
     {
-      "epoch": 2.3,
-      "eval_loss": 0.04485001042485237,
-      "eval_runtime": 237.9271,
-      "eval_samples_per_second": 81.243,
-      "eval_steps_per_second": 20.313,
-      "step": 200000
+      "epoch": 0.81,
+      "eval_loss": 0.039720822125673294,
+      "eval_runtime": 282.371,
+      "eval_samples_per_second": 78.018,
+      "eval_steps_per_second": 19.506,
+      "step": 80000
     },
     {
-      "epoch": 2.53,
-      "learning_rate": 3.735399613722064e-05,
+      "epoch": 0.91,
+      "learning_rate": 4.546073536087154e-05,
+      "loss": 0.0438,
+      "step": 90000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.49563726231906e-05,
+      "loss": 0.0438,
+      "step": 100000
+    },
+    {
+      "epoch": 1.01,
+      "eval_loss": 0.04048037528991699,
+      "eval_runtime": 267.0469,
+      "eval_samples_per_second": 82.495,
+      "eval_steps_per_second": 20.626,
+      "step": 100000
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.445200988550966e-05,
       "loss": 0.0445,
-      "step": 220000
+      "step": 110000
     },
     {
-      "epoch": 2.76,
-      "learning_rate": 3.620435942242252e-05,
-      "loss": 0.0435,
-      "step": 240000
+      "epoch": 1.21,
+      "learning_rate": 4.394764714782872e-05,
+      "loss": 0.0439,
+      "step": 120000
     },
     {
-      "epoch": 2.76,
-      "eval_loss": 0.04417626932263374,
-      "eval_runtime": 239.7656,
-      "eval_samples_per_second": 80.62,
-      "eval_steps_per_second": 20.157,
-      "step": 240000
+      "epoch": 1.21,
+      "eval_loss": 0.04071788489818573,
+      "eval_runtime": 265.1859,
+      "eval_samples_per_second": 83.074,
+      "eval_steps_per_second": 20.77,
+      "step": 120000
     },
     {
-      "epoch": 2.99,
-      "learning_rate": 3.505472270762439e-05,
-      "loss": 0.0443,
-      "step": 260000
+      "epoch": 1.31,
+      "learning_rate": 4.344328441014778e-05,
+      "loss": 0.0449,
+      "step": 130000
     },
     {
-      "epoch": 3.22,
-      "learning_rate": 3.390508599282627e-05,
-      "loss": 0.0425,
-      "step": 280000
+      "epoch": 1.41,
+      "learning_rate": 4.2938921672466844e-05,
+      "loss": 0.043,
+      "step": 140000
     },
     {
-      "epoch": 3.22,
-      "eval_loss": 0.04555518552660942,
-      "eval_runtime": 239.0043,
-      "eval_samples_per_second": 80.877,
-      "eval_steps_per_second": 20.221,
-      "step": 280000
+      "epoch": 1.41,
+      "eval_loss": 0.04017642140388489,
+      "eval_runtime": 280.2898,
+      "eval_samples_per_second": 78.597,
+      "eval_steps_per_second": 19.651,
+      "step": 140000
     },
     {
-      "epoch": 3.45,
-      "learning_rate": 3.2755449278028145e-05,
-      "loss": 0.0435,
-      "step": 300000
+      "epoch": 1.51,
+      "learning_rate": 4.24345589347859e-05,
+      "loss": 0.0448,
+      "step": 150000
     },
     {
-      "epoch": 3.68,
-      "learning_rate": 3.160581256323002e-05,
-      "loss": 0.0437,
-      "step": 320000
+      "epoch": 1.61,
+      "learning_rate": 4.193019619710496e-05,
+      "loss": 0.0434,
+      "step": 160000
     },
     {
-      "epoch": 3.68,
-      "eval_loss": 0.04474752023816109,
-      "eval_runtime": 291.4231,
-      "eval_samples_per_second": 66.33,
-      "eval_steps_per_second": 16.584,
-      "step": 320000
+      "epoch": 1.61,
+      "eval_loss": 0.03977196663618088,
+      "eval_runtime": 273.4991,
+      "eval_samples_per_second": 80.549,
+      "eval_steps_per_second": 20.139,
+      "step": 160000
     }
   ],
-  "max_steps": 869840,
+  "max_steps": 991350,
   "num_train_epochs": 10,
-  "total_flos": 2.2518483734502605e+17,
+  "total_flos": 1.169186500333609e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:981e074dd006f0b82f1adae94afef8c0b53f496530dbead237c5195367cb5167
+oid sha256:684e0b4330b8bb7c9b3e400b71095184c0853b5625598615f94e17fe26941f2b
 size 3119