add new models

.gitattributes

@@ -33,8 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
-checkpoint-*/*bim filter=lfs diff=lfs merge=lfs -text
-checkpoint-5000/training_args.bin filter=lfs diff=lfs merge=lfs -text
 checkpoint-10000/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
 checkpoint-10000/training_args.bin filter=lfs diff=lfs merge=lfs -text
-checkpoint-5000/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
+pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
+**/**/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text

checkpoint-5000/trainer_state.json

@@ -1,156 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 0.6308352258390109,
-  "global_step": 5000,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.06,
-      "learning_rate": 6.25e-05,
-      "loss": 5.1725,
-      "step": 500
-    },
-    {
-      "epoch": 0.06,
-      "eval_loss": 0.6939732432365417,
-      "eval_runtime": 5.8238,
-      "eval_samples_per_second": 17.171,
-      "eval_steps_per_second": 8.585,
-      "step": 500
-    },
-    {
-      "epoch": 0.13,
-      "learning_rate": 9.782608695652174e-05,
-      "loss": 0.5104,
-      "step": 1000
-    },
-    {
-      "epoch": 0.13,
-      "eval_loss": 0.641968846321106,
-      "eval_runtime": 6.0125,
-      "eval_samples_per_second": 16.632,
-      "eval_steps_per_second": 8.316,
-      "step": 1000
-    },
-    {
-      "epoch": 0.19,
-      "learning_rate": 9.239130434782609e-05,
-      "loss": 0.485,
-      "step": 1500
-    },
-    {
-      "epoch": 0.19,
-      "eval_loss": 0.6206984519958496,
-      "eval_runtime": 5.6411,
-      "eval_samples_per_second": 17.727,
-      "eval_steps_per_second": 8.864,
-      "step": 1500
-    },
-    {
-      "epoch": 0.25,
-      "learning_rate": 8.695652173913044e-05,
-      "loss": 0.4644,
-      "step": 2000
-    },
-    {
-      "epoch": 0.25,
-      "eval_loss": 0.6094934344291687,
-      "eval_runtime": 4.7677,
-      "eval_samples_per_second": 20.974,
-      "eval_steps_per_second": 10.487,
-      "step": 2000
-    },
-    {
-      "epoch": 0.32,
-      "learning_rate": 8.152173913043478e-05,
-      "loss": 0.4563,
-      "step": 2500
-    },
-    {
-      "epoch": 0.32,
-      "eval_loss": 0.6136000156402588,
-      "eval_runtime": 5.1919,
-      "eval_samples_per_second": 19.261,
-      "eval_steps_per_second": 9.63,
-      "step": 2500
-    },
-    {
-      "epoch": 0.38,
-      "learning_rate": 7.608695652173914e-05,
-      "loss": 0.4426,
-      "step": 3000
-    },
-    {
-      "epoch": 0.38,
-      "eval_loss": 0.6097093224525452,
-      "eval_runtime": 5.562,
-      "eval_samples_per_second": 17.979,
-      "eval_steps_per_second": 8.99,
-      "step": 3000
-    },
-    {
-      "epoch": 0.44,
-      "learning_rate": 7.065217391304349e-05,
-      "loss": 0.4401,
-      "step": 3500
-    },
-    {
-      "epoch": 0.44,
-      "eval_loss": 0.5967560410499573,
-      "eval_runtime": 5.8426,
-      "eval_samples_per_second": 17.116,
-      "eval_steps_per_second": 8.558,
-      "step": 3500
-    },
-    {
-      "epoch": 0.5,
-      "learning_rate": 6.521739130434783e-05,
-      "loss": 0.4258,
-      "step": 4000
-    },
-    {
-      "epoch": 0.5,
-      "eval_loss": 0.6082923412322998,
-      "eval_runtime": 5.3584,
-      "eval_samples_per_second": 18.662,
-      "eval_steps_per_second": 9.331,
-      "step": 4000
-    },
-    {
-      "epoch": 0.57,
-      "learning_rate": 5.9782608695652175e-05,
-      "loss": 0.424,
-      "step": 4500
-    },
-    {
-      "epoch": 0.57,
-      "eval_loss": 0.5975988507270813,
-      "eval_runtime": 4.7572,
-      "eval_samples_per_second": 21.021,
-      "eval_steps_per_second": 10.51,
-      "step": 4500
-    },
-    {
-      "epoch": 0.63,
-      "learning_rate": 5.4347826086956524e-05,
-      "loss": 0.4375,
-      "step": 5000
-    },
-    {
-      "epoch": 0.63,
-      "eval_loss": 0.5958980321884155,
-      "eval_runtime": 5.8907,
-      "eval_samples_per_second": 16.976,
-      "eval_steps_per_second": 8.488,
-      "step": 5000
-    }
-  ],
-  "max_steps": 10000,
-  "num_train_epochs": 2,
-  "total_flos": 1.921873153314816e+16,
-  "trial_name": null,
-  "trial_params": null
-}

checkpoint-5000/optimizer.pt → optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:daf86974bcdabe8d4b82019dd893671ec7dc44eebe5d722a1ca82b4ca3e2208a
+oid sha256:50d4d0984f6d3afa328cdb63469c5a8ff9da0e29f70ebc2b46f803ea273608cd
 size 2371333

checkpoint-5000/pytorch_model.bin → pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a03e000db887a4c2ae1e3911ba55550f93b8468c29500cd6a9bca46d21b96adc
+oid sha256:70f77c75f6db835162d64802678671bae5af6ec92907394913c274e72964591a
 size 990408885

checkpoint-5000/rng_state.pth → rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab9b4ed4f719b5c091aaee0f8a3b0a2f1977980e6eb7bbd803bf6300cea01e57
+oid sha256:96805bdb6c1ed875dd7da931970e8b852a1c747ef4c73dba10e101dc01ad5c13
 size 14575

checkpoint-5000/scheduler.pt → scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5db814da74fa52897e2706d0a6e369da3f074aa77a52775256fa20b47664cee
+oid sha256:3f5e24b8bf255cbefe9d307944a9741807d095b40cc5429a7befe9515b366b0f
 size 627

trainer_state.json

@@ -0,0 +1,296 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.5195263290501386,
+  "global_step": 10000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.5e-05,
+      "loss": 0.8023,
+      "step": 500
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.6604204177856445,
+      "eval_runtime": 5.6621,
+      "eval_samples_per_second": 52.984,
+      "eval_steps_per_second": 26.492,
+      "step": 500
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9e-05,
+      "loss": 0.7113,
+      "step": 1000
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 0.6501660943031311,
+      "eval_runtime": 5.6949,
+      "eval_samples_per_second": 52.679,
+      "eval_steps_per_second": 26.339,
+      "step": 1000
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 8.5e-05,
+      "loss": 0.7123,
+      "step": 1500
+    },
+    {
+      "epoch": 0.38,
+      "eval_loss": 0.6351694464683533,
+      "eval_runtime": 7.0443,
+      "eval_samples_per_second": 42.588,
+      "eval_steps_per_second": 21.294,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 8e-05,
+      "loss": 0.7002,
+      "step": 2000
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 0.6252465844154358,
+      "eval_runtime": 5.6908,
+      "eval_samples_per_second": 52.716,
+      "eval_steps_per_second": 26.358,
+      "step": 2000
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.6974,
+      "step": 2500
+    },
+    {
+      "epoch": 0.63,
+      "eval_loss": 0.6316436529159546,
+      "eval_runtime": 5.5384,
+      "eval_samples_per_second": 54.167,
+      "eval_steps_per_second": 27.083,
+      "step": 2500
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 7e-05,
+      "loss": 0.668,
+      "step": 3000
+    },
+    {
+      "epoch": 0.76,
+      "eval_loss": 0.6197097301483154,
+      "eval_runtime": 5.9147,
+      "eval_samples_per_second": 50.721,
+      "eval_steps_per_second": 25.361,
+      "step": 3000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 0.6721,
+      "step": 3500
+    },
+    {
+      "epoch": 0.88,
+      "eval_loss": 0.6143491268157959,
+      "eval_runtime": 6.0601,
+      "eval_samples_per_second": 49.504,
+      "eval_steps_per_second": 24.752,
+      "step": 3500
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6e-05,
+      "loss": 0.6603,
+      "step": 4000
+    },
+    {
+      "epoch": 1.01,
+      "eval_loss": 0.6048210859298706,
+      "eval_runtime": 5.3107,
+      "eval_samples_per_second": 56.49,
+      "eval_steps_per_second": 28.245,
+      "step": 4000
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 5.500000000000001e-05,
+      "loss": 0.598,
+      "step": 4500
+    },
+    {
+      "epoch": 1.13,
+      "eval_loss": 0.6044912934303284,
+      "eval_runtime": 5.4715,
+      "eval_samples_per_second": 54.829,
+      "eval_steps_per_second": 27.415,
+      "step": 4500
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 5e-05,
+      "loss": 0.5961,
+      "step": 5000
+    },
+    {
+      "epoch": 1.26,
+      "eval_loss": 0.5996153950691223,
+      "eval_runtime": 5.2909,
+      "eval_samples_per_second": 56.701,
+      "eval_steps_per_second": 28.35,
+      "step": 5000
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5712,
+      "step": 5500
+    },
+    {
+      "epoch": 1.39,
+      "eval_loss": 0.6034131646156311,
+      "eval_runtime": 5.0673,
+      "eval_samples_per_second": 59.203,
+      "eval_steps_per_second": 29.602,
+      "step": 5500
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 4e-05,
+      "loss": 0.5863,
+      "step": 6000
+    },
+    {
+      "epoch": 1.51,
+      "eval_loss": 0.5980101227760315,
+      "eval_runtime": 5.146,
+      "eval_samples_per_second": 58.297,
+      "eval_steps_per_second": 29.149,
+      "step": 6000
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.5e-05,
+      "loss": 0.5814,
+      "step": 6500
+    },
+    {
+      "epoch": 1.64,
+      "eval_loss": 0.5993916988372803,
+      "eval_runtime": 5.2338,
+      "eval_samples_per_second": 57.32,
+      "eval_steps_per_second": 28.66,
+      "step": 6500
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 3e-05,
+      "loss": 0.5726,
+      "step": 7000
+    },
+    {
+      "epoch": 1.76,
+      "eval_loss": 0.5940994024276733,
+      "eval_runtime": 5.1782,
+      "eval_samples_per_second": 57.935,
+      "eval_steps_per_second": 28.968,
+      "step": 7000
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 2.5e-05,
+      "loss": 0.5751,
+      "step": 7500
+    },
+    {
+      "epoch": 1.89,
+      "eval_loss": 0.5939777493476868,
+      "eval_runtime": 5.4507,
+      "eval_samples_per_second": 55.039,
+      "eval_steps_per_second": 27.52,
+      "step": 7500
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2e-05,
+      "loss": 0.5831,
+      "step": 8000
+    },
+    {
+      "epoch": 2.02,
+      "eval_loss": 0.5925641059875488,
+      "eval_runtime": 5.3244,
+      "eval_samples_per_second": 56.344,
+      "eval_steps_per_second": 28.172,
+      "step": 8000
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 1.5e-05,
+      "loss": 0.542,
+      "step": 8500
+    },
+    {
+      "epoch": 2.14,
+      "eval_loss": 0.597409188747406,
+      "eval_runtime": 5.5596,
+      "eval_samples_per_second": 53.96,
+      "eval_steps_per_second": 26.98,
+      "step": 8500
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 1e-05,
+      "loss": 0.5256,
+      "step": 9000
+    },
+    {
+      "epoch": 2.27,
+      "eval_loss": 0.594516396522522,
+      "eval_runtime": 5.486,
+      "eval_samples_per_second": 54.684,
+      "eval_steps_per_second": 27.342,
+      "step": 9000
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 5e-06,
+      "loss": 0.543,
+      "step": 9500
+    },
+    {
+      "epoch": 2.39,
+      "eval_loss": 0.592924177646637,
+      "eval_runtime": 5.3977,
+      "eval_samples_per_second": 55.579,
+      "eval_steps_per_second": 27.789,
+      "step": 9500
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 0.0,
+      "loss": 0.5272,
+      "step": 10000
+    },
+    {
+      "epoch": 2.52,
+      "eval_loss": 0.593059241771698,
+      "eval_runtime": 5.7212,
+      "eval_samples_per_second": 52.437,
+      "eval_steps_per_second": 26.218,
+      "step": 10000
+    }
+  ],
+  "max_steps": 10000,
+  "num_train_epochs": 3,
+  "total_flos": 1.2121549777083802e+17,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-5000/training_args.bin → training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92133b734236cd14623f5f32d7dac13646e59db8ee0cae12878e084143f07cb4
+oid sha256:8289752bca98a3f14e53feb07fafa691076be700549694b90e50635b318dad97
 size 4219