Training in progress, step 20000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36903f0243a52bf56c2303ead198d638e5dcf25f476cfa5723954adea34fe732
+oid sha256:74edd0cbfef63026b31ad51e8f4bc01025df6b2f2a0198b920cb1fc03a8579b7
 size 893438545

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9720327dd677676021978fd27f18fb24496429062f72218e72439f8001b20be8
+oid sha256:532ee195f2906ab9ce0bf8722baaca50a9bcc629a6a0003a6ae623a01b7ea889
 size 449471589

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c101353913b86ff575c9018e352c46e19aa450e36d2ee34d697b2d5ed877d1d
+oid sha256:b9e0931a17310d4650ee3518b83c618a6ffd8c6840ad8bd778d326fb339eb375
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c101353913b86ff575c9018e352c46e19aa450e36d2ee34d697b2d5ed877d1d
+oid sha256:b9e0931a17310d4650ee3518b83c618a6ffd8c6840ad8bd778d326fb339eb375
 size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c101353913b86ff575c9018e352c46e19aa450e36d2ee34d697b2d5ed877d1d
+oid sha256:b9e0931a17310d4650ee3518b83c618a6ffd8c6840ad8bd778d326fb339eb375
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c101353913b86ff575c9018e352c46e19aa450e36d2ee34d697b2d5ed877d1d
+oid sha256:b9e0931a17310d4650ee3518b83c618a6ffd8c6840ad8bd778d326fb339eb375
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c101353913b86ff575c9018e352c46e19aa450e36d2ee34d697b2d5ed877d1d
+oid sha256:b9e0931a17310d4650ee3518b83c618a6ffd8c6840ad8bd778d326fb339eb375
 size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c101353913b86ff575c9018e352c46e19aa450e36d2ee34d697b2d5ed877d1d
+oid sha256:b9e0931a17310d4650ee3518b83c618a6ffd8c6840ad8bd778d326fb339eb375
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c101353913b86ff575c9018e352c46e19aa450e36d2ee34d697b2d5ed877d1d
+oid sha256:b9e0931a17310d4650ee3518b83c618a6ffd8c6840ad8bd778d326fb339eb375
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c101353913b86ff575c9018e352c46e19aa450e36d2ee34d697b2d5ed877d1d
+oid sha256:b9e0931a17310d4650ee3518b83c618a6ffd8c6840ad8bd778d326fb339eb375
 size 14503

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6647dd16e16d38ed2bd48d2720f065055b2d402dd5cc12bc5f1fbd386132f2a
+oid sha256:5ca250cf2344e8df6eed15e28ea548564b75ef302cf69c48962d23fb49df8b25
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.11152872422292362,
-  "global_step": 10000,
+  "epoch": 0.22305744844584724,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -206,11 +206,211 @@
       "eval_samples_per_second": 957.11,
       "eval_steps_per_second": 15.0,
       "step": 10000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 3.149999999999999e-05,
+      "loss": 0.6293,
+      "step": 10500
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 3.2999999999999996e-05,
+      "loss": 0.6245,
+      "step": 11000
+    },
+    {
+      "epoch": 0.12,
+      "eval_loss": 0.6135886311531067,
+      "eval_runtime": 2.3072,
+      "eval_samples_per_second": 995.595,
+      "eval_steps_per_second": 15.604,
+      "step": 11000
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 3.45e-05,
+      "loss": 0.6197,
+      "step": 11500
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 3.5999999999999994e-05,
+      "loss": 0.6148,
+      "step": 12000
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.6036174297332764,
+      "eval_runtime": 2.3862,
+      "eval_samples_per_second": 962.612,
+      "eval_steps_per_second": 15.087,
+      "step": 12000
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 3.75e-05,
+      "loss": 0.6096,
+      "step": 12500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 3.9e-05,
+      "loss": 0.6046,
+      "step": 13000
+    },
+    {
+      "epoch": 0.14,
+      "eval_loss": 0.5924868583679199,
+      "eval_runtime": 2.386,
+      "eval_samples_per_second": 962.685,
+      "eval_steps_per_second": 15.088,
+      "step": 13000
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.05e-05,
+      "loss": 0.5998,
+      "step": 13500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.2e-05,
+      "loss": 0.5951,
+      "step": 14000
+    },
+    {
+      "epoch": 0.16,
+      "eval_loss": 0.5810989141464233,
+      "eval_runtime": 2.4081,
+      "eval_samples_per_second": 953.875,
+      "eval_steps_per_second": 14.95,
+      "step": 14000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.3499999999999993e-05,
+      "loss": 0.5908,
+      "step": 14500
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.4999999999999996e-05,
+      "loss": 0.586,
+      "step": 15000
+    },
+    {
+      "epoch": 0.17,
+      "eval_loss": 0.5703989863395691,
+      "eval_runtime": 2.3852,
+      "eval_samples_per_second": 963.035,
+      "eval_steps_per_second": 15.093,
+      "step": 15000
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.65e-05,
+      "loss": 0.5815,
+      "step": 15500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.7999999999999994e-05,
+      "loss": 0.5769,
+      "step": 16000
+    },
+    {
+      "epoch": 0.18,
+      "eval_loss": 0.5616013407707214,
+      "eval_runtime": 2.3872,
+      "eval_samples_per_second": 962.217,
+      "eval_steps_per_second": 15.08,
+      "step": 16000
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.95e-05,
+      "loss": 0.5722,
+      "step": 16500
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 5.1e-05,
+      "loss": 0.5673,
+      "step": 17000
+    },
+    {
+      "epoch": 0.19,
+      "eval_loss": 0.5503653287887573,
+      "eval_runtime": 2.3451,
+      "eval_samples_per_second": 979.501,
+      "eval_steps_per_second": 15.351,
+      "step": 17000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 5.2499999999999995e-05,
+      "loss": 0.5616,
+      "step": 17500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 5.399999999999999e-05,
+      "loss": 0.5553,
+      "step": 18000
+    },
+    {
+      "epoch": 0.2,
+      "eval_loss": 0.5397240519523621,
+      "eval_runtime": 2.4358,
+      "eval_samples_per_second": 943.009,
+      "eval_steps_per_second": 14.779,
+      "step": 18000
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 5.5499999999999994e-05,
+      "loss": 0.5491,
+      "step": 18500
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 5.6999999999999996e-05,
+      "loss": 0.5421,
+      "step": 19000
+    },
+    {
+      "epoch": 0.21,
+      "eval_loss": 0.5266169905662537,
+      "eval_runtime": 2.4081,
+      "eval_samples_per_second": 953.853,
+      "eval_steps_per_second": 14.949,
+      "step": 19000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 5.85e-05,
+      "loss": 0.5357,
+      "step": 19500
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 5.9999999999999995e-05,
+      "loss": 0.5301,
+      "step": 20000
+    },
+    {
+      "epoch": 0.22,
+      "eval_loss": 0.5150080919265747,
+      "eval_runtime": 2.3725,
+      "eval_samples_per_second": 968.19,
+      "eval_steps_per_second": 15.174,
+      "step": 20000
     }
   ],
   "max_steps": 1000000,
   "num_train_epochs": 12,
-  "total_flos": 7.010026103034715e+20,
+  "total_flos": 1.402005220606943e+21,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9720327dd677676021978fd27f18fb24496429062f72218e72439f8001b20be8
+oid sha256:532ee195f2906ab9ce0bf8722baaca50a9bcc629a6a0003a6ae623a01b7ea889
 size 449471589