Training in progress, step 10000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f559368effc2bcb9ed41744d242e8a6179a8a3b02ba25e13e1f8450bb4bc3e4
+oid sha256:ec52caad045b3f34632491a3e60266fca49fb4a8996857ded69858dac82d39b4
 size 202193937

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a261c0cd0888dbefe81cc7be4d2ade2ab4eef0a9d6eb7436783fa2a5af2f7c9
+oid sha256:c4934ec353ee8c0b201d8896e01193b73d8d525ed3d2c04d990e1d6af1e4f1ff
 size 102501541

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69f5d31aae33dba6f6e60bfe4edf45a94d300c4e16fb452118b96c312aa296ee
+oid sha256:161364992086d421e640cb46d16fb2f26cefdd719c2e2fd276cf5231abbcc82c
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69f5d31aae33dba6f6e60bfe4edf45a94d300c4e16fb452118b96c312aa296ee
+oid sha256:161364992086d421e640cb46d16fb2f26cefdd719c2e2fd276cf5231abbcc82c
 size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69f5d31aae33dba6f6e60bfe4edf45a94d300c4e16fb452118b96c312aa296ee
+oid sha256:161364992086d421e640cb46d16fb2f26cefdd719c2e2fd276cf5231abbcc82c
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69f5d31aae33dba6f6e60bfe4edf45a94d300c4e16fb452118b96c312aa296ee
+oid sha256:161364992086d421e640cb46d16fb2f26cefdd719c2e2fd276cf5231abbcc82c
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69f5d31aae33dba6f6e60bfe4edf45a94d300c4e16fb452118b96c312aa296ee
+oid sha256:161364992086d421e640cb46d16fb2f26cefdd719c2e2fd276cf5231abbcc82c
 size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69f5d31aae33dba6f6e60bfe4edf45a94d300c4e16fb452118b96c312aa296ee
+oid sha256:161364992086d421e640cb46d16fb2f26cefdd719c2e2fd276cf5231abbcc82c
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69f5d31aae33dba6f6e60bfe4edf45a94d300c4e16fb452118b96c312aa296ee
+oid sha256:161364992086d421e640cb46d16fb2f26cefdd719c2e2fd276cf5231abbcc82c
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69f5d31aae33dba6f6e60bfe4edf45a94d300c4e16fb452118b96c312aa296ee
+oid sha256:161364992086d421e640cb46d16fb2f26cefdd719c2e2fd276cf5231abbcc82c
 size 14503

last-checkpoint/text_renderer_config.json

@@ -16,18 +16,20 @@
     "Nimbus Roman",
     "Nimbus Sans",
     "Nimbus Sans Narrow",
+    "Noto Fangsong KSS Rotated",
+    "Noto Fangsong KSS Vertical",
     "Noto Kufi Arabic",
-    "Noto Looped Lao",
-    "Noto Looped Lao UI",
     "Noto Music",
     "Noto Naskh Arabic",
     "Noto Naskh Arabic UI",
     "Noto Nastaliq Urdu",
     "Noto Rashi Hebrew",
+    "Noto Sans",
     "Noto Sans Adlam",
     "Noto Sans Adlam Unjoined",
     "Noto Sans Anatolian Hieroglyphs",
     "Noto Sans Arabic",
+    "Noto Sans Arabic UI",
     "Noto Sans Armenian",
     "Noto Sans Avestan",
     "Noto Sans Balinese",
@@ -86,12 +88,16 @@
     "Noto Sans Kaithi",
     "Noto Sans Kannada",
     "Noto Sans Kannada UI",
+    "Noto Sans Kawi",
     "Noto Sans Kayah Li",
     "Noto Sans Kharoshthi",
     "Noto Sans Khmer",
     "Noto Sans Khmer UI",
+    "Noto Sans Khojki",
     "Noto Sans Khudawadi",
     "Noto Sans Lao",
+    "Noto Sans Lao Looped",
+    "Noto Sans Lao Looped UI",
     "Noto Sans Lao UI",
     "Noto Sans Lepcha",
     "Noto Sans Limbu",
@@ -123,13 +129,15 @@
     "Noto Sans Myanmar UI",
     "Noto Sans NKo",
     "Noto Sans Nabataean",
+    "Noto Sans Nag Mundari",
     "Noto Sans Nandinagari",
     "Noto Sans New Tai Lue",
     "Noto Sans Newa",
     "Noto Sans Nushu",
     "Noto Sans Ogham",
-    "Noto Sans Old",
+    "Noto Sans Ol Chiki",
     "Noto Sans Old Hungarian",
+    "Noto Sans Old Italic",
     "Noto Sans Old North Arabian",
     "Noto Sans Old Permic",
     "Noto Sans Old Persian",
@@ -142,7 +150,7 @@
     "Noto Sans Pahawh Hmong",
     "Noto Sans Palmyrene",
     "Noto Sans Pau Cin Hau",
-    "Noto Sans PhagsPa",
+    "Noto Sans Phags-Pa",
     "Noto Sans Phoenician",
     "Noto Sans Psalter Pahlavi",
     "Noto Sans Rejang",
@@ -156,11 +164,15 @@
     "Noto Sans Sinhala",
     "Noto Sans Sinhala UI",
     "Noto Sans Sogdian",
+    "Noto Sans Sora Sompeng",
     "Noto Sans Soyombo",
+    "Noto Sans Sundanese",
     "Noto Sans Syloti Nagri",
     "Noto Sans Symbols",
-    "Noto Sans Symbols2",
+    "Noto Sans Symbols 2",
     "Noto Sans Syriac",
+    "Noto Sans Syriac Eastern",
+    "Noto Sans Syriac Western",
     "Noto Sans Tagalog",
     "Noto Sans Tagbanwa",
     "Noto Sans Tai Le",

last-checkpoint/trainer_state.json

@@ -1,216 +1,216 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.061849357011635,
+  "epoch": 0.22305496074232692,
   "global_step": 10000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.15,
+      "epoch": 0.01,
       "learning_rate": 5.999999999999999e-06,
-      "loss": 0.9314,
+      "loss": 0.9165,
       "step": 500
     },
     {
-      "epoch": 0.31,
+      "epoch": 0.02,
       "learning_rate": 1.1999999999999999e-05,
-      "loss": 0.7106,
+      "loss": 0.7051,
       "step": 1000
     },
     {
-      "epoch": 0.31,
-      "eval_loss": 0.6879013180732727,
-      "eval_runtime": 1.244,
-      "eval_samples_per_second": 135.052,
-      "eval_steps_per_second": 2.412,
+      "epoch": 0.02,
+      "eval_loss": 0.6818633079528809,
+      "eval_runtime": 1.9324,
+      "eval_samples_per_second": 1188.697,
+      "eval_steps_per_second": 18.63,
       "step": 1000
     },
     {
-      "epoch": 0.46,
+      "epoch": 0.03,
       "learning_rate": 1.7999999999999997e-05,
-      "loss": 0.6869,
+      "loss": 0.6818,
       "step": 1500
     },
     {
-      "epoch": 0.61,
+      "epoch": 0.04,
       "learning_rate": 2.3999999999999997e-05,
-      "loss": 0.6858,
+      "loss": 0.6808,
       "step": 2000
     },
     {
-      "epoch": 0.61,
-      "eval_loss": 0.6865440607070923,
-      "eval_runtime": 1.2991,
-      "eval_samples_per_second": 129.324,
-      "eval_steps_per_second": 2.309,
+      "epoch": 0.04,
+      "eval_loss": 0.6788448691368103,
+      "eval_runtime": 1.762,
+      "eval_samples_per_second": 1303.62,
+      "eval_steps_per_second": 20.431,
       "step": 2000
     },
     {
-      "epoch": 0.77,
+      "epoch": 0.06,
       "learning_rate": 2.9999999999999997e-05,
-      "loss": 0.6856,
+      "loss": 0.6804,
       "step": 2500
     },
     {
-      "epoch": 0.92,
+      "epoch": 0.07,
       "learning_rate": 3.5999999999999994e-05,
-      "loss": 0.6853,
+      "loss": 0.6803,
       "step": 3000
     },
     {
-      "epoch": 0.92,
-      "eval_loss": 0.6834043860435486,
-      "eval_runtime": 1.2199,
-      "eval_samples_per_second": 137.717,
-      "eval_steps_per_second": 2.459,
+      "epoch": 0.07,
+      "eval_loss": 0.678871750831604,
+      "eval_runtime": 1.9035,
+      "eval_samples_per_second": 1206.735,
+      "eval_steps_per_second": 18.913,
       "step": 3000
     },
     {
-      "epoch": 1.07,
+      "epoch": 0.08,
       "learning_rate": 4.2e-05,
-      "loss": 0.6852,
+      "loss": 0.6801,
       "step": 3500
     },
     {
-      "epoch": 1.22,
+      "epoch": 0.09,
       "learning_rate": 4.7999999999999994e-05,
-      "loss": 0.685,
+      "loss": 0.68,
       "step": 4000
     },
     {
-      "epoch": 1.22,
-      "eval_loss": 0.6872440576553345,
-      "eval_runtime": 1.1829,
-      "eval_samples_per_second": 142.026,
-      "eval_steps_per_second": 2.536,
+      "epoch": 0.09,
+      "eval_loss": 0.6781899333000183,
+      "eval_runtime": 1.9139,
+      "eval_samples_per_second": 1200.176,
+      "eval_steps_per_second": 18.81,
       "step": 4000
     },
     {
-      "epoch": 1.38,
+      "epoch": 0.1,
       "learning_rate": 5.399999999999999e-05,
-      "loss": 0.6846,
+      "loss": 0.6798,
       "step": 4500
     },
     {
-      "epoch": 1.53,
+      "epoch": 0.11,
       "learning_rate": 5.9999999999999995e-05,
-      "loss": 0.6843,
+      "loss": 0.6796,
       "step": 5000
     },
     {
-      "epoch": 1.53,
-      "eval_loss": 0.6845670342445374,
-      "eval_runtime": 1.2484,
-      "eval_samples_per_second": 134.569,
-      "eval_steps_per_second": 2.403,
+      "epoch": 0.11,
+      "eval_loss": 0.6784412860870361,
+      "eval_runtime": 1.8311,
+      "eval_samples_per_second": 1254.405,
+      "eval_steps_per_second": 19.66,
       "step": 5000
     },
     {
-      "epoch": 1.68,
+      "epoch": 0.12,
       "learning_rate": 6.599999999999999e-05,
-      "loss": 0.6838,
+      "loss": 0.6792,
       "step": 5500
     },
     {
-      "epoch": 1.84,
+      "epoch": 0.13,
       "learning_rate": 7.199999999999999e-05,
-      "loss": 0.6833,
+      "loss": 0.6788,
       "step": 6000
     },
     {
-      "epoch": 1.84,
-      "eval_loss": 0.685808539390564,
-      "eval_runtime": 1.254,
-      "eval_samples_per_second": 133.975,
-      "eval_steps_per_second": 2.392,
+      "epoch": 0.13,
+      "eval_loss": 0.67804354429245,
+      "eval_runtime": 1.8509,
+      "eval_samples_per_second": 1240.986,
+      "eval_steps_per_second": 19.449,
       "step": 6000
     },
     {
-      "epoch": 1.99,
+      "epoch": 0.14,
       "learning_rate": 7.8e-05,
-      "loss": 0.6807,
+      "loss": 0.6783,
       "step": 6500
     },
     {
-      "epoch": 2.14,
+      "epoch": 0.16,
       "learning_rate": 8.4e-05,
-      "loss": 0.6623,
+      "loss": 0.678,
       "step": 7000
     },
     {
-      "epoch": 2.14,
-      "eval_loss": 0.6461814045906067,
-      "eval_runtime": 1.2227,
-      "eval_samples_per_second": 137.404,
-      "eval_steps_per_second": 2.454,
+      "epoch": 0.16,
+      "eval_loss": 0.6776260733604431,
+      "eval_runtime": 1.8893,
+      "eval_samples_per_second": 1215.825,
+      "eval_steps_per_second": 19.055,
       "step": 7000
     },
     {
-      "epoch": 2.3,
+      "epoch": 0.17,
       "learning_rate": 8.999999999999999e-05,
-      "loss": 0.646,
+      "loss": 0.6778,
       "step": 7500
     },
     {
-      "epoch": 2.45,
+      "epoch": 0.18,
       "learning_rate": 9.599999999999999e-05,
-      "loss": 0.6376,
+      "loss": 0.678,
       "step": 8000
     },
     {
-      "epoch": 2.45,
-      "eval_loss": 0.6246757507324219,
-      "eval_runtime": 1.3234,
-      "eval_samples_per_second": 126.944,
-      "eval_steps_per_second": 2.267,
+      "epoch": 0.18,
+      "eval_loss": 0.6782737374305725,
+      "eval_runtime": 1.8707,
+      "eval_samples_per_second": 1227.889,
+      "eval_steps_per_second": 19.244,
       "step": 8000
     },
     {
-      "epoch": 2.6,
+      "epoch": 0.19,
       "learning_rate": 0.000102,
-      "loss": 0.6291,
+      "loss": 0.6783,
       "step": 8500
     },
     {
-      "epoch": 2.76,
+      "epoch": 0.2,
       "learning_rate": 0.00010799999999999998,
-      "loss": 0.6186,
+      "loss": 0.6672,
       "step": 9000
     },
     {
-      "epoch": 2.76,
-      "eval_loss": 0.6001520156860352,
-      "eval_runtime": 1.2491,
-      "eval_samples_per_second": 134.495,
-      "eval_steps_per_second": 2.402,
+      "epoch": 0.2,
+      "eval_loss": 0.6526092290878296,
+      "eval_runtime": 1.8141,
+      "eval_samples_per_second": 1266.219,
+      "eval_steps_per_second": 19.845,
       "step": 9000
     },
     {
-      "epoch": 2.91,
+      "epoch": 0.21,
       "learning_rate": 0.00011399999999999999,
-      "loss": 0.6061,
+      "loss": 0.6453,
       "step": 9500
     },
     {
-      "epoch": 3.06,
+      "epoch": 0.22,
       "learning_rate": 0.00011999999999999999,
-      "loss": 0.5917,
+      "loss": 0.632,
       "step": 10000
     },
     {
-      "epoch": 3.06,
-      "eval_loss": 0.5758978724479675,
-      "eval_runtime": 1.1861,
-      "eval_samples_per_second": 141.636,
-      "eval_steps_per_second": 2.529,
+      "epoch": 0.22,
+      "eval_loss": 0.6212196350097656,
+      "eval_runtime": 1.9366,
+      "eval_samples_per_second": 1186.109,
+      "eval_steps_per_second": 18.589,
       "step": 10000
     }
   ],
   "max_steps": 500000,
-  "num_train_epochs": 154,
-  "total_flos": 3.194571847792245e+20,
+  "num_train_epochs": 12,
+  "total_flos": 3.194871387745e+20,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d73ef3420d099bbd958456da5993d0c834d0afa3b0922d167ad1cdd0b4a7760
+oid sha256:229efc1d49e061de5d19e6e496114be212e29dd2150b027ea46959c2a5a56c0b
 size 3311

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a261c0cd0888dbefe81cc7be4d2ade2ab4eef0a9d6eb7436783fa2a5af2f7c9
+oid sha256:c4934ec353ee8c0b201d8896e01193b73d8d525ed3d2c04d990e1d6af1e4f1ff
 size 102501541

text_renderer_config.json

@@ -16,18 +16,20 @@
     "Nimbus Roman",
     "Nimbus Sans",
     "Nimbus Sans Narrow",
+    "Noto Fangsong KSS Rotated",
+    "Noto Fangsong KSS Vertical",
     "Noto Kufi Arabic",
-    "Noto Looped Lao",
-    "Noto Looped Lao UI",
     "Noto Music",
     "Noto Naskh Arabic",
     "Noto Naskh Arabic UI",
     "Noto Nastaliq Urdu",
     "Noto Rashi Hebrew",
+    "Noto Sans",
     "Noto Sans Adlam",
     "Noto Sans Adlam Unjoined",
     "Noto Sans Anatolian Hieroglyphs",
     "Noto Sans Arabic",
+    "Noto Sans Arabic UI",
     "Noto Sans Armenian",
     "Noto Sans Avestan",
     "Noto Sans Balinese",
@@ -86,12 +88,16 @@
     "Noto Sans Kaithi",
     "Noto Sans Kannada",
     "Noto Sans Kannada UI",
+    "Noto Sans Kawi",
     "Noto Sans Kayah Li",
     "Noto Sans Kharoshthi",
     "Noto Sans Khmer",
     "Noto Sans Khmer UI",
+    "Noto Sans Khojki",
     "Noto Sans Khudawadi",
     "Noto Sans Lao",
+    "Noto Sans Lao Looped",
+    "Noto Sans Lao Looped UI",
     "Noto Sans Lao UI",
     "Noto Sans Lepcha",
     "Noto Sans Limbu",
@@ -123,13 +129,15 @@
     "Noto Sans Myanmar UI",
     "Noto Sans NKo",
     "Noto Sans Nabataean",
+    "Noto Sans Nag Mundari",
     "Noto Sans Nandinagari",
     "Noto Sans New Tai Lue",
     "Noto Sans Newa",
     "Noto Sans Nushu",
     "Noto Sans Ogham",
-    "Noto Sans Old",
+    "Noto Sans Ol Chiki",
     "Noto Sans Old Hungarian",
+    "Noto Sans Old Italic",
     "Noto Sans Old North Arabian",
     "Noto Sans Old Permic",
     "Noto Sans Old Persian",
@@ -142,7 +150,7 @@
     "Noto Sans Pahawh Hmong",
     "Noto Sans Palmyrene",
     "Noto Sans Pau Cin Hau",
-    "Noto Sans PhagsPa",
+    "Noto Sans Phags-Pa",
     "Noto Sans Phoenician",
     "Noto Sans Psalter Pahlavi",
     "Noto Sans Rejang",
@@ -156,11 +164,15 @@
     "Noto Sans Sinhala",
     "Noto Sans Sinhala UI",
     "Noto Sans Sogdian",
+    "Noto Sans Sora Sompeng",
     "Noto Sans Soyombo",
+    "Noto Sans Sundanese",
     "Noto Sans Syloti Nagri",
     "Noto Sans Symbols",
-    "Noto Sans Symbols2",
+    "Noto Sans Symbols 2",
     "Noto Sans Syriac",
+    "Noto Sans Syriac Eastern",
+    "Noto Sans Syriac Western",
     "Noto Sans Tagalog",
     "Noto Sans Tagbanwa",
     "Noto Sans Tai Le",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d73ef3420d099bbd958456da5993d0c834d0afa3b0922d167ad1cdd0b4a7760
+oid sha256:229efc1d49e061de5d19e6e496114be212e29dd2150b027ea46959c2a5a56c0b
 size 3311