Model save

README.md

@@ -2,15 +2,11 @@
 license: mit
 base_model: gpt2
 tags:
-- alignment-handbook
-- trl
-- sft
-- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- appliedai-qx/sample-dataset-ah
+- generator
 model-index:
 - name: gpt2
   results: []
@@ -21,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gpt2
 
-This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the appliedai-qx/sample-dataset-ah dataset.
+This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
 
 ## Model description
 

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 7357983621120000.0,
-    "train_loss": 1.386462221362374,
-    "train_runtime": 54.2438,
+    "train_loss": 1.3830752210183577,
+    "train_runtime": 55.0309,
     "train_samples": 10000,
-    "train_samples_per_second": 258.942,
-    "train_steps_per_second": 4.056
+    "train_samples_per_second": 255.239,
+    "train_steps_per_second": 3.998
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be7733f81180fdd29d3314b1b3412be39b7b836f121522629c20b72e54edbb5d
+oid sha256:c41f0cbc1c377cca8babf7afd81c7f49cfaff8ba4c2bb995d16ba8207d5a008c
 size 248894656

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 7357983621120000.0,
-    "train_loss": 1.386462221362374,
-    "train_runtime": 54.2438,
+    "train_loss": 1.3830752210183577,
+    "train_runtime": 55.0309,
     "train_samples": 10000,
-    "train_samples_per_second": 258.942,
-    "train_steps_per_second": 4.056
+    "train_samples_per_second": 255.239,
+    "train_steps_per_second": 3.998
 }

trainer_state.json

@@ -17,320 +17,320 @@
     },
     {
       "epoch": 0.022727272727272728,
-      "grad_norm": 3.28125,
+      "grad_norm": 3.296875,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 1.872,
+      "loss": 1.8721,
       "step": 5
     },
     {
       "epoch": 0.045454545454545456,
       "grad_norm": 1.7578125,
       "learning_rate": 9.090909090909092e-05,
-      "loss": 1.8049,
+      "loss": 1.8044,
       "step": 10
     },
     {
       "epoch": 0.06818181818181818,
       "grad_norm": 1.4375,
       "learning_rate": 0.00013636363636363637,
-      "loss": 1.74,
+      "loss": 1.7402,
       "step": 15
     },
     {
       "epoch": 0.09090909090909091,
-      "grad_norm": 2.0,
+      "grad_norm": 1.9921875,
       "learning_rate": 0.00018181818181818183,
-      "loss": 1.6831,
+      "loss": 1.6832,
       "step": 20
     },
     {
       "epoch": 0.11363636363636363,
-      "grad_norm": 1.609375,
+      "grad_norm": 1.6640625,
       "learning_rate": 0.0001998867339183008,
-      "loss": 1.607,
+      "loss": 1.6069,
       "step": 25
     },
     {
       "epoch": 0.13636363636363635,
-      "grad_norm": 1.6796875,
+      "grad_norm": 1.4609375,
       "learning_rate": 0.00019919548128307954,
-      "loss": 1.5824,
+      "loss": 1.582,
       "step": 30
     },
     {
       "epoch": 0.1590909090909091,
-      "grad_norm": 1.25,
+      "grad_norm": 1.453125,
       "learning_rate": 0.00019788024462147788,
-      "loss": 1.5335,
+      "loss": 1.5337,
       "step": 35
     },
     {
       "epoch": 0.18181818181818182,
-      "grad_norm": 0.9453125,
+      "grad_norm": 0.9140625,
       "learning_rate": 0.00019594929736144976,
-      "loss": 1.4989,
+      "loss": 1.4999,
       "step": 40
     },
     {
       "epoch": 0.20454545454545456,
-      "grad_norm": 1.5546875,
+      "grad_norm": 0.91796875,
       "learning_rate": 0.00019341478602651069,
-      "loss": 1.4829,
+      "loss": 1.486,
       "step": 45
     },
     {
       "epoch": 0.22727272727272727,
-      "grad_norm": 1.3203125,
+      "grad_norm": 0.9140625,
       "learning_rate": 0.00019029265382866214,
-      "loss": 1.4675,
+      "loss": 1.4694,
       "step": 50
     },
     {
       "epoch": 0.25,
-      "grad_norm": 1.3828125,
+      "grad_norm": 0.8125,
       "learning_rate": 0.00018660254037844388,
-      "loss": 1.4425,
+      "loss": 1.4435,
       "step": 55
     },
     {
       "epoch": 0.2727272727272727,
-      "grad_norm": 1.2890625,
+      "grad_norm": 0.85546875,
       "learning_rate": 0.0001823676581429833,
-      "loss": 1.4339,
+      "loss": 1.4326,
       "step": 60
     },
     {
       "epoch": 0.29545454545454547,
-      "grad_norm": 0.7890625,
+      "grad_norm": 0.82421875,
       "learning_rate": 0.0001776146464291757,
-      "loss": 1.4086,
+      "loss": 1.4055,
       "step": 65
     },
     {
       "epoch": 0.3181818181818182,
-      "grad_norm": 0.7421875,
+      "grad_norm": 0.8046875,
       "learning_rate": 0.00017237340381050703,
-      "loss": 1.3789,
+      "loss": 1.374,
       "step": 70
     },
     {
       "epoch": 0.3409090909090909,
-      "grad_norm": 0.71484375,
+      "grad_norm": 0.96484375,
       "learning_rate": 0.00016667690005162916,
-      "loss": 1.3825,
+      "loss": 1.3776,
       "step": 75
     },
     {
       "epoch": 0.36363636363636365,
-      "grad_norm": 0.6484375,
+      "grad_norm": 0.71875,
       "learning_rate": 0.00016056096871376667,
-      "loss": 1.3689,
+      "loss": 1.3626,
       "step": 80
     },
     {
       "epoch": 0.38636363636363635,
-      "grad_norm": 0.671875,
+      "grad_norm": 0.6328125,
       "learning_rate": 0.00015406408174555976,
-      "loss": 1.363,
+      "loss": 1.3566,
       "step": 85
     },
     {
       "epoch": 0.4090909090909091,
-      "grad_norm": 0.68359375,
+      "grad_norm": 0.72265625,
       "learning_rate": 0.0001472271074772683,
-      "loss": 1.3277,
+      "loss": 1.3215,
       "step": 90
     },
     {
       "epoch": 0.4318181818181818,
-      "grad_norm": 0.7890625,
+      "grad_norm": 0.82421875,
       "learning_rate": 0.00014009305354066137,
-      "loss": 1.3438,
+      "loss": 1.338,
       "step": 95
     },
     {
       "epoch": 0.45454545454545453,
-      "grad_norm": 0.8125,
+      "grad_norm": 0.87109375,
       "learning_rate": 0.00013270679633174218,
-      "loss": 1.3325,
+      "loss": 1.3268,
       "step": 100
     },
     {
       "epoch": 0.4772727272727273,
-      "grad_norm": 0.9921875,
+      "grad_norm": 0.78515625,
       "learning_rate": 0.0001251147987181079,
-      "loss": 1.3127,
+      "loss": 1.307,
       "step": 105
     },
     {
       "epoch": 0.5,
-      "grad_norm": 0.82421875,
+      "grad_norm": 1.03125,
       "learning_rate": 0.00011736481776669306,
-      "loss": 1.3142,
+      "loss": 1.3086,
       "step": 110
     },
     {
       "epoch": 0.5227272727272727,
-      "grad_norm": 0.7109375,
+      "grad_norm": 0.609375,
       "learning_rate": 0.00010950560433041826,
-      "loss": 1.3031,
+      "loss": 1.2977,
       "step": 115
     },
     {
       "epoch": 0.5454545454545454,
-      "grad_norm": 0.67578125,
+      "grad_norm": 0.65625,
       "learning_rate": 0.00010158659638348081,
-      "loss": 1.3119,
+      "loss": 1.3071,
       "step": 120
     },
     {
       "epoch": 0.5681818181818182,
       "grad_norm": 0.5859375,
       "learning_rate": 9.365760803434355e-05,
-      "loss": 1.3096,
+      "loss": 1.305,
       "step": 125
     },
     {
       "epoch": 0.5909090909090909,
-      "grad_norm": 0.52734375,
+      "grad_norm": 0.55859375,
       "learning_rate": 8.57685161726715e-05,
-      "loss": 1.2951,
+      "loss": 1.2902,
       "step": 130
     },
     {
       "epoch": 0.6136363636363636,
-      "grad_norm": 0.5234375,
+      "grad_norm": 0.51953125,
       "learning_rate": 7.796894672134594e-05,
-      "loss": 1.2856,
+      "loss": 1.281,
       "step": 135
     },
     {
       "epoch": 0.6363636363636364,
-      "grad_norm": 0.69921875,
+      "grad_norm": 0.66796875,
       "learning_rate": 7.030796246717255e-05,
-      "loss": 1.3007,
+      "loss": 1.296,
       "step": 140
     },
     {
       "epoch": 0.6590909090909091,
-      "grad_norm": 0.67578125,
+      "grad_norm": 0.6875,
       "learning_rate": 6.283375443396726e-05,
-      "loss": 1.293,
+      "loss": 1.2883,
       "step": 145
     },
     {
       "epoch": 0.6818181818181818,
-      "grad_norm": 0.6015625,
+      "grad_norm": 0.62890625,
       "learning_rate": 5.559333873942259e-05,
-      "loss": 1.2909,
+      "loss": 1.2866,
       "step": 150
     },
     {
       "epoch": 0.7045454545454546,
-      "grad_norm": 0.7265625,
+      "grad_norm": 0.74609375,
       "learning_rate": 4.8632260842659393e-05,
-      "loss": 1.2965,
+      "loss": 1.2921,
       "step": 155
     },
     {
       "epoch": 0.7272727272727273,
-      "grad_norm": 0.51953125,
+      "grad_norm": 0.53125,
       "learning_rate": 4.19943090428802e-05,
-      "loss": 1.2818,
+      "loss": 1.2774,
       "step": 160
     },
     {
       "epoch": 0.75,
-      "grad_norm": 0.451171875,
+      "grad_norm": 0.458984375,
       "learning_rate": 3.5721239031346066e-05,
-      "loss": 1.2593,
+      "loss": 1.2547,
       "step": 165
     },
     {
       "epoch": 0.7727272727272727,
-      "grad_norm": 0.494140625,
+      "grad_norm": 0.515625,
       "learning_rate": 2.9852511229367865e-05,
-      "loss": 1.2859,
+      "loss": 1.2814,
       "step": 170
     },
     {
       "epoch": 0.7954545454545454,
-      "grad_norm": 0.439453125,
+      "grad_norm": 0.453125,
       "learning_rate": 2.4425042564574184e-05,
-      "loss": 1.2837,
+      "loss": 1.2796,
       "step": 175
     },
     {
       "epoch": 0.8181818181818182,
-      "grad_norm": 0.44140625,
+      "grad_norm": 0.455078125,
       "learning_rate": 1.947297424689414e-05,
-      "loss": 1.2852,
+      "loss": 1.281,
       "step": 180
     },
     {
       "epoch": 0.8409090909090909,
-      "grad_norm": 0.447265625,
+      "grad_norm": 0.45703125,
       "learning_rate": 1.5027457005048573e-05,
-      "loss": 1.2796,
+      "loss": 1.2752,
       "step": 185
     },
     {
       "epoch": 0.8636363636363636,
-      "grad_norm": 0.66796875,
+      "grad_norm": 0.69140625,
       "learning_rate": 1.1116455134507664e-05,
-      "loss": 1.2792,
+      "loss": 1.2744,
       "step": 190
     },
     {
       "epoch": 0.8863636363636364,
-      "grad_norm": 0.44921875,
+      "grad_norm": 0.453125,
       "learning_rate": 7.764570589541875e-06,
-      "loss": 1.28,
+      "loss": 1.2756,
       "step": 195
     },
     {
       "epoch": 0.9090909090909091,
-      "grad_norm": 0.470703125,
+      "grad_norm": 0.47265625,
       "learning_rate": 4.992888225905468e-06,
-      "loss": 1.2861,
+      "loss": 1.2816,
       "step": 200
     },
     {
       "epoch": 0.9318181818181818,
-      "grad_norm": 0.546875,
+      "grad_norm": 0.55859375,
       "learning_rate": 2.818843167645835e-06,
-      "loss": 1.2822,
+      "loss": 1.278,
       "step": 205
     },
     {
       "epoch": 0.9545454545454546,
-      "grad_norm": 0.470703125,
+      "grad_norm": 0.478515625,
       "learning_rate": 1.2561111323605712e-06,
-      "loss": 1.28,
+      "loss": 1.2754,
       "step": 210
     },
     {
       "epoch": 0.9772727272727273,
-      "grad_norm": 0.462890625,
+      "grad_norm": 0.45703125,
       "learning_rate": 3.145224048057727e-07,
-      "loss": 1.2803,
+      "loss": 1.2759,
       "step": 215
     },
     {
       "epoch": 1.0,
-      "grad_norm": 0.47265625,
+      "grad_norm": 0.478515625,
       "learning_rate": 0.0,
-      "loss": 1.2746,
+      "loss": 1.2702,
       "step": 220
     },
     {
       "epoch": 1.0,
       "step": 220,
       "total_flos": 7357983621120000.0,
-      "train_loss": 1.386462221362374,
-      "train_runtime": 54.2438,
-      "train_samples_per_second": 258.942,
-      "train_steps_per_second": 4.056
+      "train_loss": 1.3830752210183577,
+      "train_runtime": 55.0309,
+      "train_samples_per_second": 255.239,
+      "train_steps_per_second": 3.998
     }
   ],
   "logging_steps": 5,