medquad

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7eefbc60c75eef82bd84155bf83b4cbc42ebe86ca062e3169bb4b11bda454bea
+oid sha256:d680bb0c04772bf8f307ff155fdada2c78e6740d40f29e832d239e9f0eca7ebf
 size 19644664

all_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 10.0,
-    "total_flos": 854477494026240.0,
-    "train_loss": 2.395888161659241,
-    "train_runtime": 85.4261,
-    "train_samples": 6174,
-    "train_samples_per_second": 0.936,
-    "train_steps_per_second": 0.117
+    "epoch": 5.0,
+    "total_flos": 549306960445440.0,
+    "train_loss": 2.4234995365142824,
+    "train_runtime": 54.6732,
+    "train_samples": 8643,
+    "train_samples_per_second": 1.463,
+    "train_steps_per_second": 0.183
 }

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:547c0e52df3d3df3f60b259eb50cfe3bc5175623ef3ad49023ee4287dbd617ea
-size 34315473
+oid sha256:950330f025e1c5f06810c89ffeac976f405aa09b24aadddcb699dd6067a28407
+size 34315572

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 10.0,
-    "total_flos": 854477494026240.0,
-    "train_loss": 2.395888161659241,
-    "train_runtime": 85.4261,
-    "train_samples": 6174,
-    "train_samples_per_second": 0.936,
-    "train_steps_per_second": 0.117
+    "epoch": 5.0,
+    "total_flos": 549306960445440.0,
+    "train_loss": 2.4234995365142824,
+    "train_runtime": 54.6732,
+    "train_samples": 8643,
+    "train_samples_per_second": 1.463,
+    "train_steps_per_second": 0.183
 }

trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 5.0,
   "eval_steps": 10,
   "global_step": 10,
   "is_hyper_param_search": false,
@@ -9,163 +9,123 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "grad_norm": 3.581932544708252,
+      "epoch": 0.8,
+      "grad_norm": 3.535655975341797,
       "learning_rate": 0.0001,
-      "loss": 2.8417,
+      "loss": 2.8223,
       "step": 1
     },
     {
       "epoch": 1.0,
-      "eval_loss": 2.8694076538085938,
-      "eval_runtime": 0.5128,
-      "eval_samples_per_second": 1.95,
-      "eval_steps_per_second": 1.95,
-      "step": 1
-    },
-    {
-      "epoch": 2.0,
-      "grad_norm": 3.5751430988311768,
+      "grad_norm": 4.1554155349731445,
       "learning_rate": 0.0002,
-      "loss": 2.8416,
+      "loss": 2.8268,
       "step": 2
     },
     {
-      "epoch": 2.0,
-      "eval_loss": 2.695298671722412,
-      "eval_runtime": 0.5118,
-      "eval_samples_per_second": 1.954,
-      "eval_steps_per_second": 1.954,
+      "epoch": 1.0,
+      "eval_loss": 2.803156852722168,
+      "eval_runtime": 0.5136,
+      "eval_samples_per_second": 1.947,
+      "eval_steps_per_second": 1.947,
       "step": 2
     },
     {
-      "epoch": 3.0,
-      "grad_norm": 2.5421221256256104,
+      "epoch": 1.8,
+      "grad_norm": 2.6138558387756348,
       "learning_rate": 0.000175,
-      "loss": 2.6806,
-      "step": 3
-    },
-    {
-      "epoch": 3.0,
-      "eval_loss": 2.450862169265747,
-      "eval_runtime": 0.5063,
-      "eval_samples_per_second": 1.975,
-      "eval_steps_per_second": 1.975,
+      "loss": 2.6803,
       "step": 3
     },
     {
-      "epoch": 4.0,
-      "grad_norm": 1.2186226844787598,
+      "epoch": 2.0,
+      "grad_norm": 2.1446619033813477,
       "learning_rate": 0.00015000000000000001,
-      "loss": 2.4573,
+      "loss": 2.4443,
       "step": 4
     },
     {
-      "epoch": 4.0,
-      "eval_loss": 2.3435235023498535,
-      "eval_runtime": 0.5052,
-      "eval_samples_per_second": 1.979,
-      "eval_steps_per_second": 1.979,
+      "epoch": 2.0,
+      "eval_loss": 2.5913567543029785,
+      "eval_runtime": 0.5175,
+      "eval_samples_per_second": 1.932,
+      "eval_steps_per_second": 1.932,
       "step": 4
     },
     {
-      "epoch": 5.0,
-      "grad_norm": 1.7033910751342773,
+      "epoch": 2.8,
+      "grad_norm": 1.7912520170211792,
       "learning_rate": 0.000125,
-      "loss": 2.3686,
-      "step": 5
-    },
-    {
-      "epoch": 5.0,
-      "eval_loss": 2.263502597808838,
-      "eval_runtime": 0.5027,
-      "eval_samples_per_second": 1.989,
-      "eval_steps_per_second": 1.989,
+      "loss": 2.4006,
       "step": 5
     },
     {
-      "epoch": 6.0,
-      "grad_norm": 1.8212623596191406,
+      "epoch": 3.0,
+      "grad_norm": 2.2134900093078613,
       "learning_rate": 0.0001,
-      "loss": 2.2829,
+      "loss": 2.3568,
       "step": 6
     },
     {
-      "epoch": 6.0,
-      "eval_loss": 2.1960206031799316,
-      "eval_runtime": 0.5032,
-      "eval_samples_per_second": 1.987,
-      "eval_steps_per_second": 1.987,
+      "epoch": 3.0,
+      "eval_loss": 2.456364870071411,
+      "eval_runtime": 0.5073,
+      "eval_samples_per_second": 1.971,
+      "eval_steps_per_second": 1.971,
       "step": 6
     },
     {
-      "epoch": 7.0,
-      "grad_norm": 1.5627810955047607,
+      "epoch": 3.8,
+      "grad_norm": 1.7705835103988647,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 2.1995,
+      "loss": 2.2664,
       "step": 7
     },
     {
-      "epoch": 7.0,
-      "eval_loss": 2.146212339401245,
-      "eval_runtime": 0.5043,
-      "eval_samples_per_second": 1.983,
-      "eval_steps_per_second": 1.983,
-      "step": 7
-    },
-    {
-      "epoch": 8.0,
-      "grad_norm": 1.2493813037872314,
+      "epoch": 4.0,
+      "grad_norm": 1.7635867595672607,
       "learning_rate": 5e-05,
-      "loss": 2.1334,
+      "loss": 2.1515,
       "step": 8
     },
     {
-      "epoch": 8.0,
-      "eval_loss": 2.1154565811157227,
-      "eval_runtime": 0.5079,
-      "eval_samples_per_second": 1.969,
-      "eval_steps_per_second": 1.969,
+      "epoch": 4.0,
+      "eval_loss": 2.3617074489593506,
+      "eval_runtime": 0.5027,
+      "eval_samples_per_second": 1.989,
+      "eval_steps_per_second": 1.989,
       "step": 8
     },
     {
-      "epoch": 9.0,
-      "grad_norm": 1.068291187286377,
+      "epoch": 4.8,
+      "grad_norm": 1.2085267305374146,
       "learning_rate": 2.5e-05,
-      "loss": 2.0905,
-      "step": 9
-    },
-    {
-      "epoch": 9.0,
-      "eval_loss": 2.0988733768463135,
-      "eval_runtime": 0.5081,
-      "eval_samples_per_second": 1.968,
-      "eval_steps_per_second": 1.968,
+      "loss": 2.1536,
       "step": 9
     },
     {
-      "epoch": 10.0,
-      "grad_norm": 1.0751439332962036,
+      "epoch": 5.0,
+      "grad_norm": 1.5746992826461792,
       "learning_rate": 0.0,
-      "loss": 2.0628,
+      "loss": 2.1324,
       "step": 10
     },
     {
-      "epoch": 10.0,
-      "eval_loss": 2.0916833877563477,
-      "eval_runtime": 0.5125,
-      "eval_samples_per_second": 1.951,
-      "eval_steps_per_second": 1.951,
+      "epoch": 5.0,
+      "eval_loss": 2.3257529735565186,
+      "eval_runtime": 0.5048,
+      "eval_samples_per_second": 1.981,
+      "eval_steps_per_second": 1.981,
       "step": 10
     },
     {
-      "epoch": 10.0,
+      "epoch": 5.0,
       "step": 10,
-      "total_flos": 854477494026240.0,
-      "train_loss": 2.395888161659241,
-      "train_runtime": 85.4261,
-      "train_samples_per_second": 0.936,
-      "train_steps_per_second": 0.117
+      "total_flos": 549306960445440.0,
+      "train_loss": 2.4234995365142824,
+      "train_runtime": 54.6732,
+      "train_samples_per_second": 1.463,
+      "train_steps_per_second": 0.183
     }
   ],
   "logging_steps": 1,
@@ -185,7 +145,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 854477494026240.0,
+  "total_flos": 549306960445440.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:630d77759ac46171dd0558f612e4e399f4a265d0d4e2d95dba9f52ec5936b894
+oid sha256:2b6901fa5e2bec60b8edebb9df49acc61f121566b10b9726cffffd490adbbb3e
 size 5624