Training in progress, step 2500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:746732a955ce5ac447aae4ae74e4f595331f971b2b8b78db1b28ff7cc415b7b9
+oid sha256:2cf50cbd134a3066e0d74687b39b2bba0fc2a0cdb34238e44082bdf70216c841
 size 268290900

run-12/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d0c54d41e69626eaa1d38e2df7307881acf7d56b84fa8cd48c73da5bab91925
+oid sha256:9920b10690143b73412fcfb36733c8bedabcfc46ba38a96125fc271c0c292eb2
 size 268290900

run-12/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf2af8af45e5e5addf8b25086866db7509d70ff6a036d63377dd3109ba6e22eb
+oid sha256:4494feaf7fb5dd7e1d0fc07621e5798771a3885406d9de8742ca5224611675c4
 size 536643898

run-12/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c794bc4c67ef18245dd516031ce405ab557e4d551d225d8dd1e1abc0f2be8e33
+oid sha256:72a4307cca951ef6d42c7f184774e8b1259c78b2bf130b8443ec0b62f6dbf689
 size 1064

run-12/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec0cc6acf298edb76f3b5ee73f8edb2020abda4adeb81e0f2c668de6a46acc2f
+oid sha256:825ce3dae72879e90f1e7c8bf9db5978e39cf9b4e9cf4d01acbdb15f1075ac03
 size 5176

run-12/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42a45feb81fc68209cef779ccddc8347fdc246602423d8ee79e52c992b11b094
+oid sha256:6b75f5a956ae846daba9cd3186cf441c4ca189b39bd62da8a093ac836d1f5f70
 size 268290900

run-12/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:280bd9284ea1c8b42ee189fab53af2cdb70f15b6985d9d1c1c8a692032dc0e8e
+oid sha256:4abbf43eef188aa2e0d4cb960d43a14b2b15884f4c3a6a3cc53ff412ef763a6b
 size 536643898

run-12/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71453465aad25f4c5a0a948496c64b1f74df850abda497954afe3695c00756ee
+oid sha256:693ec081eb0fa0af7d13e289edb194450dd19fdada23346e2af4292ea228535f
 size 1064

run-12/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec0cc6acf298edb76f3b5ee73f8edb2020abda4adeb81e0f2c668de6a46acc2f
+oid sha256:825ce3dae72879e90f1e7c8bf9db5978e39cf9b4e9cf4d01acbdb15f1075ac03
 size 5176

run-12/checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:435a53c8e54107bdc0570aa20eaee5791887a3630f7c40ff4f94645fe82c714c
+oid sha256:232eaf6924b640670e424f121b0308a7c44eba675a1105736c179b9e656b4fc8
 size 268290900

run-12/checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c5bb6be470b5c7d27395588f256ac99dc123e4e5ddbfb2a18b7cd4981e6cadf
+oid sha256:7b52890dbff2fde64fbafc70d437485424d61246dc71300b50df3dcbeb702a8a
 size 536643898

run-12/checkpoint-2000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43d45918bfeb622ba899798234ec7646c7a90dd3f5771db086f5b7dee1d5a530
+oid sha256:67d1a8a9215f18c672cf798b531337b9e0636922d8c4373e76080f8fec19ceb2
 size 1064

run-12/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec0cc6acf298edb76f3b5ee73f8edb2020abda4adeb81e0f2c668de6a46acc2f
+oid sha256:825ce3dae72879e90f1e7c8bf9db5978e39cf9b4e9cf4d01acbdb15f1075ac03
 size 5176

run-12/checkpoint-2500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5562e40e2d8c4c5b6d75eaadaf13c35e3662d9300701698371ddeb320098f37a
+oid sha256:2cf50cbd134a3066e0d74687b39b2bba0fc2a0cdb34238e44082bdf70216c841
 size 268290900

run-12/checkpoint-2500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbd728c822f7cc7ad773da4adc5666e4c902e7852e11797fe28e4c72c37bfafb
+oid sha256:0fac8f5ce1a1a762a78650e0ad4773833a79e3d301a367a8ab899a2adfbd9404
 size 536643898

run-12/checkpoint-2500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d116f7be47ce88249a782ce7aafaf5b673a05bfa0ebdf9f14ab4c327c94c6b04
+oid sha256:b7c943a77651a184c6653dd4706c14c9a27d52136e81942a2b361c74acb9f665
 size 1064

run-12/checkpoint-2500/trainer_state.json

@@ -10,107 +10,107 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6064516129032258,
-      "eval_loss": 0.2185761034488678,
-      "eval_runtime": 1.3638,
-      "eval_samples_per_second": 2273.102,
-      "eval_steps_per_second": 47.662,
+      "eval_accuracy": 0.6732258064516129,
+      "eval_loss": 0.40791815519332886,
+      "eval_runtime": 1.3596,
+      "eval_samples_per_second": 2280.098,
+      "eval_steps_per_second": 47.809,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5516161918640137,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3519,
+      "grad_norm": 1.0064228773117065,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.642,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8412903225806452,
-      "eval_loss": 0.10163559764623642,
-      "eval_runtime": 1.3673,
-      "eval_samples_per_second": 2267.277,
-      "eval_steps_per_second": 47.54,
+      "eval_accuracy": 0.8425806451612903,
+      "eval_loss": 0.14076808094978333,
+      "eval_runtime": 1.3551,
+      "eval_samples_per_second": 2287.635,
+      "eval_steps_per_second": 47.967,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8858064516129033,
-      "eval_loss": 0.06496834009885788,
-      "eval_runtime": 1.3574,
-      "eval_samples_per_second": 2283.817,
-      "eval_steps_per_second": 47.886,
+      "eval_accuracy": 0.8961290322580645,
+      "eval_loss": 0.07251902669668198,
+      "eval_runtime": 1.3507,
+      "eval_samples_per_second": 2295.17,
+      "eval_steps_per_second": 48.125,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.5089602470397949,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1182,
+      "grad_norm": 0.7264623045921326,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1644,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9003225806451612,
-      "eval_loss": 0.048449017107486725,
-      "eval_runtime": 1.3674,
-      "eval_samples_per_second": 2267.023,
-      "eval_steps_per_second": 47.534,
+      "eval_accuracy": 0.9141935483870968,
+      "eval_loss": 0.05225619301199913,
+      "eval_runtime": 1.3588,
+      "eval_samples_per_second": 2281.449,
+      "eval_steps_per_second": 47.837,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.3282025456428528,
-      "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.0719,
+      "grad_norm": 0.4414461851119995,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.0818,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9135483870967742,
-      "eval_loss": 0.03883416950702667,
-      "eval_runtime": 1.3721,
-      "eval_samples_per_second": 2259.288,
-      "eval_steps_per_second": 47.372,
+      "eval_accuracy": 0.9238709677419354,
+      "eval_loss": 0.04304172471165657,
+      "eval_runtime": 1.3623,
+      "eval_samples_per_second": 2275.546,
+      "eval_steps_per_second": 47.713,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.922258064516129,
-      "eval_loss": 0.03377517685294151,
-      "eval_runtime": 1.5655,
-      "eval_samples_per_second": 1980.228,
-      "eval_steps_per_second": 41.521,
+      "eval_accuracy": 0.9283870967741935,
+      "eval_loss": 0.038618117570877075,
+      "eval_runtime": 1.3565,
+      "eval_samples_per_second": 2285.293,
+      "eval_steps_per_second": 47.917,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.24476511776447296,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0554,
+      "grad_norm": 0.3635008931159973,
+      "learning_rate": 6.02375960866527e-06,
+      "loss": 0.0619,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9251612903225807,
-      "eval_loss": 0.03079158626496792,
-      "eval_runtime": 1.3605,
-      "eval_samples_per_second": 2278.646,
-      "eval_steps_per_second": 47.778,
+      "eval_accuracy": 0.9316129032258065,
+      "eval_loss": 0.03583395853638649,
+      "eval_runtime": 1.356,
+      "eval_samples_per_second": 2286.159,
+      "eval_steps_per_second": 47.936,
       "step": 2226
     },
     {
       "epoch": 7.861635220125786,
-      "grad_norm": 0.23761987686157227,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0476,
+      "grad_norm": 0.3274884819984436,
+      "learning_rate": 2.5296995108315863e-06,
+      "loss": 0.0541,
       "step": 2500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -128,8 +128,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.86409290852114,
-    "num_train_epochs": 10,
-    "temperature": 6
+    "alpha": 0.9838674034859954,
+    "num_train_epochs": 9,
+    "temperature": 2
   }
 }

run-12/checkpoint-2500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec0cc6acf298edb76f3b5ee73f8edb2020abda4adeb81e0f2c668de6a46acc2f
+oid sha256:825ce3dae72879e90f1e7c8bf9db5978e39cf9b4e9cf4d01acbdb15f1075ac03
 size 5176