Model save

README.md

@@ -2,10 +2,11 @@
 license: apache-2.0
 base_model: mistralai/Mistral-7B-v0.1
 tags:
-- alignment-handbook
+- trl
+- sft
 - generated_from_trainer
 datasets:
-- HuggingFaceH4/ultrachat_200k
+- generator
 model-index:
 - name: zephyr-7b-sft-full
   results: []
@@ -16,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-sft-full
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the HuggingFaceH4/ultrachat_200k dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9353
+- Loss: 0.9354
 
 ## Model description
 
@@ -54,7 +55,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.9078        | 1.0   | 1090 | 0.9353          |
+| 0.9077        | 1.0   | 1090 | 0.9354          |
 
 
 ### Framework versions

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 1.0,
-    "eval_loss": 0.9352946281433105,
-    "eval_runtime": 142.3405,
+    "eval_loss": 0.9353834986686707,
+    "eval_runtime": 143.2861,
     "eval_samples": 23110,
-    "eval_samples_per_second": 108.409,
-    "eval_steps_per_second": 1.7,
-    "train_loss": 0.07620820999145508,
-    "train_runtime": 563.6077,
+    "eval_samples_per_second": 107.694,
+    "eval_steps_per_second": 1.689,
+    "train_loss": 0.9776670060026537,
+    "train_runtime": 5614.3798,
     "train_samples": 207865,
-    "train_samples_per_second": 247.381,
-    "train_steps_per_second": 1.934
+    "train_samples_per_second": 24.834,
+    "train_steps_per_second": 0.194
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_loss": 0.9352946281433105,
-    "eval_runtime": 142.3405,
+    "eval_loss": 0.9353834986686707,
+    "eval_runtime": 143.2861,
     "eval_samples": 23110,
-    "eval_samples_per_second": 108.409,
-    "eval_steps_per_second": 1.7
+    "eval_samples_per_second": 107.694,
+    "eval_steps_per_second": 1.689
 }

model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ad03c68e305895320be8556507a5c7fe172be4c848b327e7154482a0e051600
+oid sha256:c8f0995a437e42c9046c3964bb7d858c34866d6b4f1de4ef12f98129888ee3fc
 size 4943162336

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3442a9f89309cb48f3c120d06169c98cf42e0f0bb185ad262da6eaf5b65e217d
+oid sha256:3b789d79949a4f5de8001f9e01a619675b6d2675041e49fb912ebe6c8e7d5b52
 size 4999819336

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a5a61a77bbe6ccdcc60aacbb76e657e45c1a17ea23a8fd93bf7fe3ff861d5ad
+oid sha256:4d4b611a06358f2462ff5cc487cd28bd2de498fde53ba13e219b972869ebe1d5
 size 4540516344

runs/Jan09_22-11-21_ip-26-0-162-180/events.out.tfevents.1704838311.ip-26-0-162-180.3647178.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85f9255ae3dff40993d24ef4a650264c1b83ab8a8fc1b2fc4b6998ad82e94d76
-size 35929
+oid sha256:d28c38c3c3a2d5d09592b34f1885d240ced6b061000d86207dd6fce4a2a33095
+size 39380

runs/Jan09_22-11-21_ip-26-0-162-180/events.out.tfevents.1704844068.ip-26-0-162-180.3647178.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cfef66c0ba02c9e54feb1537c8fc054d48605df617e900c782d113fa2cc0859
+size 359

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "train_loss": 0.07620820999145508,
-    "train_runtime": 563.6077,
+    "train_loss": 0.9776670060026537,
+    "train_runtime": 5614.3798,
     "train_samples": 207865,
-    "train_samples_per_second": 247.381,
-    "train_steps_per_second": 1.934
+    "train_samples_per_second": 24.834,
+    "train_steps_per_second": 0.194
 }

trainer_state.json

@@ -23,7 +23,7 @@
     {
       "epoch": 0.01,
       "learning_rate": 1.8348623853211011e-06,
-      "loss": 1.3919,
+      "loss": 1.3918,
       "step": 10
     },
     {
@@ -35,13 +35,13 @@
     {
       "epoch": 0.02,
       "learning_rate": 3.6697247706422022e-06,
-      "loss": 1.1699,
+      "loss": 1.17,
       "step": 20
     },
     {
       "epoch": 0.02,
       "learning_rate": 4.587155963302753e-06,
-      "loss": 1.1172,
+      "loss": 1.1173,
       "step": 25
     },
     {
@@ -53,13 +53,13 @@
     {
       "epoch": 0.03,
       "learning_rate": 6.422018348623854e-06,
-      "loss": 1.0575,
+      "loss": 1.0576,
       "step": 35
     },
     {
       "epoch": 0.04,
       "learning_rate": 7.3394495412844045e-06,
-      "loss": 1.052,
+      "loss": 1.0521,
       "step": 40
     },
     {
@@ -77,7 +77,7 @@
     {
       "epoch": 0.05,
       "learning_rate": 1.0091743119266055e-05,
-      "loss": 1.0091,
+      "loss": 1.0092,
       "step": 55
     },
     {
@@ -107,31 +107,31 @@
     {
       "epoch": 0.07,
       "learning_rate": 1.4678899082568809e-05,
-      "loss": 0.9919,
+      "loss": 0.992,
       "step": 80
     },
     {
       "epoch": 0.08,
       "learning_rate": 1.559633027522936e-05,
-      "loss": 1.0149,
+      "loss": 1.0151,
       "step": 85
     },
     {
       "epoch": 0.08,
       "learning_rate": 1.6513761467889912e-05,
-      "loss": 0.9972,
+      "loss": 0.9973,
       "step": 90
     },
     {
       "epoch": 0.09,
       "learning_rate": 1.743119266055046e-05,
-      "loss": 1.0152,
+      "loss": 1.0154,
       "step": 95
     },
     {
       "epoch": 0.09,
       "learning_rate": 1.834862385321101e-05,
-      "loss": 1.0163,
+      "loss": 1.0162,
       "step": 100
     },
     {
@@ -143,7 +143,7 @@
     {
       "epoch": 0.1,
       "learning_rate": 1.999994872196626e-05,
-      "loss": 1.0055,
+      "loss": 1.0054,
       "step": 110
     },
     {
@@ -155,25 +155,25 @@
     {
       "epoch": 0.11,
       "learning_rate": 1.999379599421534e-05,
-      "loss": 1.0057,
+      "loss": 1.0058,
       "step": 120
     },
     {
       "epoch": 0.11,
       "learning_rate": 1.9986875683942535e-05,
-      "loss": 1.02,
+      "loss": 1.0201,
       "step": 125
     },
     {
       "epoch": 0.12,
       "learning_rate": 1.9977394889447526e-05,
-      "loss": 1.0235,
+      "loss": 1.0236,
       "step": 130
     },
     {
       "epoch": 0.12,
       "learning_rate": 1.9965356041462954e-05,
-      "loss": 1.0302,
+      "loss": 1.0303,
       "step": 135
     },
     {
@@ -185,31 +185,31 @@
     {
       "epoch": 0.13,
       "learning_rate": 1.9933617186395917e-05,
-      "loss": 1.0045,
+      "loss": 1.0044,
       "step": 145
     },
     {
       "epoch": 0.14,
       "learning_rate": 1.9913925316676946e-05,
-      "loss": 1.0028,
+      "loss": 1.0025,
       "step": 150
     },
     {
       "epoch": 0.14,
       "learning_rate": 1.9891691666109112e-05,
-      "loss": 1.0225,
+      "loss": 1.0226,
       "step": 155
     },
     {
       "epoch": 0.15,
       "learning_rate": 1.9866921935064907e-05,
-      "loss": 1.0133,
+      "loss": 1.0136,
       "step": 160
     },
     {
       "epoch": 0.15,
       "learning_rate": 1.9839622474129595e-05,
-      "loss": 1.0159,
+      "loss": 1.0161,
       "step": 165
     },
     {
@@ -221,19 +221,19 @@
     {
       "epoch": 0.16,
       "learning_rate": 1.977746300605507e-05,
-      "loss": 1.0214,
+      "loss": 1.0223,
       "step": 175
     },
     {
       "epoch": 0.17,
       "learning_rate": 1.9742618935665478e-05,
-      "loss": 1.0084,
+      "loss": 1.0088,
       "step": 180
     },
     {
       "epoch": 0.17,
       "learning_rate": 1.9705277004798072e-05,
-      "loss": 0.9985,
+      "loss": 0.9986,
       "step": 185
     },
     {
@@ -245,43 +245,43 @@
     {
       "epoch": 0.18,
       "learning_rate": 1.9623138495219292e-05,
-      "loss": 1.0087,
+      "loss": 1.0088,
       "step": 195
     },
     {
       "epoch": 0.18,
       "learning_rate": 1.957836297558229e-05,
-      "loss": 1.0276,
+      "loss": 1.0274,
       "step": 200
     },
     {
       "epoch": 0.19,
       "learning_rate": 1.9531131708217005e-05,
-      "loss": 0.9895,
+      "loss": 0.9892,
       "step": 205
     },
     {
       "epoch": 0.19,
       "learning_rate": 1.948145680250766e-05,
-      "loss": 1.0119,
+      "loss": 1.0117,
       "step": 210
     },
     {
       "epoch": 0.2,
       "learning_rate": 1.9429350994350483e-05,
-      "loss": 0.9934,
+      "loss": 0.9933,
       "step": 215
     },
     {
       "epoch": 0.2,
       "learning_rate": 1.93748276428884e-05,
-      "loss": 1.0233,
+      "loss": 1.0232,
       "step": 220
     },
     {
       "epoch": 0.21,
       "learning_rate": 1.931790072708596e-05,
-      "loss": 1.019,
+      "loss": 1.0189,
       "step": 225
     },
     {
@@ -293,43 +293,43 @@
     {
       "epoch": 0.22,
       "learning_rate": 1.9196895195764363e-05,
-      "loss": 0.9959,
+      "loss": 0.9957,
       "step": 235
     },
     {
       "epoch": 0.22,
       "learning_rate": 1.913284760423745e-05,
-      "loss": 1.0215,
+      "loss": 1.0213,
       "step": 240
     },
     {
       "epoch": 0.22,
       "learning_rate": 1.9066458488400586e-05,
-      "loss": 1.0257,
+      "loss": 1.0258,
       "step": 245
     },
     {
       "epoch": 0.23,
       "learning_rate": 1.8997744869421248e-05,
-      "loss": 1.0038,
+      "loss": 1.0037,
       "step": 250
     },
     {
       "epoch": 0.23,
       "learning_rate": 1.8926724364434447e-05,
-      "loss": 0.9954,
+      "loss": 0.9953,
       "step": 255
     },
     {
       "epoch": 0.24,
       "learning_rate": 1.8853415182025953e-05,
-      "loss": 1.0087,
+      "loss": 1.0088,
       "step": 260
     },
     {
       "epoch": 0.24,
       "learning_rate": 1.8777836117563894e-05,
-      "loss": 1.0197,
+      "loss": 1.0198,
       "step": 265
     },
     {
@@ -341,13 +341,13 @@
     {
       "epoch": 0.25,
       "learning_rate": 1.861994642880105e-05,
-      "loss": 1.0037,
+      "loss": 1.0035,
       "step": 275
     },
     {
       "epoch": 0.26,
       "learning_rate": 1.8537676285033886e-05,
-      "loss": 1.0063,
+      "loss": 1.0062,
       "step": 280
     },
     {
@@ -359,13 +359,13 @@
     {
       "epoch": 0.27,
       "learning_rate": 1.8366590857437182e-05,
-      "loss": 1.0155,
+      "loss": 1.015,
       "step": 290
     },
     {
       "epoch": 0.27,
       "learning_rate": 1.8277819437329577e-05,
-      "loss": 0.9929,
+      "loss": 0.9928,
       "step": 295
     },
     {
@@ -377,7 +377,7 @@
     {
       "epoch": 0.28,
       "learning_rate": 1.809393297692334e-05,
-      "loss": 0.989,
+      "loss": 0.9891,
       "step": 305
     },
     {
@@ -389,25 +389,25 @@
     {
       "epoch": 0.29,
       "learning_rate": 1.790174639944997e-05,
-      "loss": 0.9892,
+      "loss": 0.9891,
       "step": 315
     },
     {
       "epoch": 0.29,
       "learning_rate": 1.780260182802314e-05,
-      "loss": 0.9981,
+      "loss": 0.998,
       "step": 320
     },
     {
       "epoch": 0.3,
       "learning_rate": 1.7701456787229805e-05,
-      "loss": 0.9873,
+      "loss": 0.9875,
       "step": 325
     },
     {
       "epoch": 0.3,
       "learning_rate": 1.7598337209132142e-05,
-      "loss": 1.0099,
+      "loss": 1.01,
       "step": 330
     },
     {
@@ -419,19 +419,19 @@
     {
       "epoch": 0.31,
       "learning_rate": 1.738628069370195e-05,
-      "loss": 0.9978,
+      "loss": 0.9979,
       "step": 340
     },
     {
       "epoch": 0.32,
       "learning_rate": 1.7277398124460022e-05,
-      "loss": 1.007,
+      "loss": 1.0069,
       "step": 345
     },
     {
       "epoch": 0.32,
       "learning_rate": 1.71666497401558e-05,
-      "loss": 0.9925,
+      "loss": 0.9926,
       "step": 350
     },
     {
@@ -443,7 +443,7 @@
     {
       "epoch": 0.33,
       "learning_rate": 1.6939669574305565e-05,
-      "loss": 1.0069,
+      "loss": 1.0071,
       "step": 360
     },
     {
@@ -455,73 +455,73 @@
     {
       "epoch": 0.34,
       "learning_rate": 1.6705572958388576e-05,
-      "loss": 0.9989,
+      "loss": 0.999,
       "step": 370
     },
     {
       "epoch": 0.34,
       "learning_rate": 1.6585930722009602e-05,
-      "loss": 0.9952,
+      "loss": 0.9953,
       "step": 375
     },
     {
       "epoch": 0.35,
       "learning_rate": 1.6464599952375998e-05,
-      "loss": 0.9971,
+      "loss": 0.9973,
       "step": 380
     },
     {
       "epoch": 0.35,
       "learning_rate": 1.63416117568662e-05,
-      "loss": 0.9812,
+      "loss": 0.9813,
       "step": 385
     },
     {
       "epoch": 0.36,
       "learning_rate": 1.621699766779763e-05,
-      "loss": 1.0093,
+      "loss": 1.009,
       "step": 390
     },
     {
       "epoch": 0.36,
       "learning_rate": 1.6090789634342278e-05,
-      "loss": 1.0029,
+      "loss": 1.003,
       "step": 395
     },
     {
       "epoch": 0.37,
       "learning_rate": 1.5963020014335437e-05,
-      "loss": 0.9931,
+      "loss": 0.9932,
       "step": 400
     },
     {
       "epoch": 0.37,
       "learning_rate": 1.583372156597961e-05,
-      "loss": 0.9753,
+      "loss": 0.9754,
       "step": 405
     },
     {
       "epoch": 0.38,
       "learning_rate": 1.570292743944583e-05,
-      "loss": 1.0094,
+      "loss": 1.0097,
       "step": 410
     },
     {
       "epoch": 0.38,
       "learning_rate": 1.557067116837444e-05,
-      "loss": 0.968,
+      "loss": 0.9678,
       "step": 415
     },
     {
       "epoch": 0.39,
       "learning_rate": 1.5436986661277578e-05,
-      "loss": 0.996,
+      "loss": 0.9961,
       "step": 420
     },
     {
       "epoch": 0.39,
       "learning_rate": 1.530190819284555e-05,
-      "loss": 1.0131,
+      "loss": 1.0128,
       "step": 425
     },
     {
@@ -533,13 +533,13 @@
     {
       "epoch": 0.4,
       "learning_rate": 1.5027708248811331e-05,
-      "loss": 0.9889,
+      "loss": 0.9888,
       "step": 435
     },
     {
       "epoch": 0.4,
       "learning_rate": 1.4888657073937077e-05,
-      "loss": 0.9789,
+      "loss": 0.979,
       "step": 440
     },
     {
@@ -551,13 +551,13 @@
     {
       "epoch": 0.41,
       "learning_rate": 1.4606830562448692e-05,
-      "loss": 0.9876,
+      "loss": 0.9875,
       "step": 450
     },
     {
       "epoch": 0.42,
       "learning_rate": 1.4464127481899312e-05,
-      "loss": 0.9998,
+      "loss": 0.9999,
       "step": 455
     },
     {
@@ -569,13 +569,13 @@
     {
       "epoch": 0.43,
       "learning_rate": 1.4175324596392075e-05,
-      "loss": 0.983,
+      "loss": 0.9832,
       "step": 465
     },
     {
       "epoch": 0.43,
       "learning_rate": 1.402929883613599e-05,
-      "loss": 0.9765,
+      "loss": 0.9766,
       "step": 470
     },
     {
@@ -587,7 +587,7 @@
     {
       "epoch": 0.44,
       "learning_rate": 1.3734185865027061e-05,
-      "loss": 0.9834,
+      "loss": 0.9833,
       "step": 480
     },
     {
@@ -599,25 +599,25 @@
     {
       "epoch": 0.45,
       "learning_rate": 1.3435243583752294e-05,
-      "loss": 0.9751,
+      "loss": 0.9752,
       "step": 490
     },
     {
       "epoch": 0.45,
       "learning_rate": 1.3284432106201233e-05,
-      "loss": 0.9798,
+      "loss": 0.9796,
       "step": 495
     },
     {
       "epoch": 0.46,
       "learning_rate": 1.313277854982062e-05,
-      "loss": 0.958,
+      "loss": 0.9584,
       "step": 500
     },
     {
       "epoch": 0.46,
       "learning_rate": 1.2980321796293838e-05,
-      "loss": 0.9886,
+      "loss": 0.9884,
       "step": 505
     },
     {
@@ -635,61 +635,61 @@
     {
       "epoch": 0.48,
       "learning_rate": 1.2518524198405699e-05,
-      "loss": 0.9661,
+      "loss": 0.9662,
       "step": 520
     },
     {
       "epoch": 0.48,
       "learning_rate": 1.2363247441059775e-05,
-      "loss": 0.9711,
+      "loss": 0.971,
       "step": 525
     },
     {
       "epoch": 0.49,
       "learning_rate": 1.2207364782731657e-05,
-      "loss": 0.9796,
+      "loss": 0.9798,
       "step": 530
     },
     {
       "epoch": 0.49,
       "learning_rate": 1.2050916189382646e-05,
-      "loss": 0.9637,
+      "loss": 0.9638,
       "step": 535
     },
     {
       "epoch": 0.5,
       "learning_rate": 1.189394177207125e-05,
-      "loss": 0.9827,
+      "loss": 0.9825,
       "step": 540
     },
     {
       "epoch": 0.5,
       "learning_rate": 1.1736481776669307e-05,
-      "loss": 0.9817,
+      "loss": 0.9816,
       "step": 545
     },
     {
       "epoch": 0.5,
       "learning_rate": 1.1578576573543541e-05,
-      "loss": 0.9613,
+      "loss": 0.9612,
       "step": 550
     },
     {
       "epoch": 0.51,
       "learning_rate": 1.1420266647205232e-05,
-      "loss": 0.9693,
+      "loss": 0.9692,
       "step": 555
     },
     {
       "epoch": 0.51,
       "learning_rate": 1.1261592585930576e-05,
-      "loss": 0.9877,
+      "loss": 0.9876,
       "step": 560
     },
     {
       "epoch": 0.52,
       "learning_rate": 1.1102595071354471e-05,
-      "loss": 0.9448,
+      "loss": 0.9446,
       "step": 565
     },
     {
@@ -701,13 +701,13 @@
     {
       "epoch": 0.53,
       "learning_rate": 1.0783792813028828e-05,
-      "loss": 0.9777,
+      "loss": 0.9776,
       "step": 575
     },
     {
       "epoch": 0.53,
       "learning_rate": 1.0624069805367558e-05,
-      "loss": 0.9401,
+      "loss": 0.9402,
       "step": 580
     },
     {
@@ -719,49 +719,49 @@
     {
       "epoch": 0.54,
       "learning_rate": 1.0304184775393642e-05,
-      "loss": 0.9572,
+      "loss": 0.9571,
       "step": 590
     },
     {
       "epoch": 0.55,
       "learning_rate": 1.0144104766775574e-05,
-      "loss": 0.9692,
+      "loss": 0.9694,
       "step": 595
     },
     {
       "epoch": 0.55,
       "learning_rate": 9.983987811869863e-06,
-      "loss": 0.9668,
+      "loss": 0.967,
       "step": 600
     },
     {
       "epoch": 0.56,
       "learning_rate": 9.823874962247565e-06,
-      "loss": 0.9654,
+      "loss": 0.9651,
       "step": 605
     },
     {
       "epoch": 0.56,
       "learning_rate": 9.663807268427197e-06,
-      "loss": 0.951,
+      "loss": 0.9511,
       "step": 610
     },
     {
       "epoch": 0.56,
       "learning_rate": 9.503825769350016e-06,
-      "loss": 0.9735,
+      "loss": 0.9731,
       "step": 615
     },
     {
       "epoch": 0.57,
       "learning_rate": 9.343971481858246e-06,
-      "loss": 0.949,
+      "loss": 0.9487,
       "step": 620
     },
     {
       "epoch": 0.57,
       "learning_rate": 9.184285390178978e-06,
-      "loss": 0.9638,
+      "loss": 0.9637,
       "step": 625
     },
     {
@@ -773,7 +773,7 @@
     {
       "epoch": 0.58,
       "learning_rate": 8.865581505055292e-06,
-      "loss": 0.9499,
+      "loss": 0.9501,
       "step": 635
     },
     {
@@ -785,133 +785,133 @@
     {
       "epoch": 0.59,
       "learning_rate": 8.548040936496989e-06,
-      "loss": 0.9454,
+      "loss": 0.9452,
       "step": 645
     },
     {
       "epoch": 0.6,
       "learning_rate": 8.389808710909881e-06,
-      "loss": 0.9738,
+      "loss": 0.9739,
       "step": 650
     },
     {
       "epoch": 0.6,
       "learning_rate": 8.231989314071318e-06,
-      "loss": 0.9609,
+      "loss": 0.9608,
       "step": 655
     },
     {
       "epoch": 0.61,
       "learning_rate": 8.07462320849313e-06,
-      "loss": 0.9601,
+      "loss": 0.9602,
       "step": 660
     },
     {
       "epoch": 0.61,
       "learning_rate": 7.917750740470116e-06,
-      "loss": 0.9694,
+      "loss": 0.9695,
       "step": 665
     },
     {
       "epoch": 0.61,
       "learning_rate": 7.761412129735853e-06,
-      "loss": 0.9625,
+      "loss": 0.9626,
       "step": 670
     },
     {
       "epoch": 0.62,
       "learning_rate": 7.605647459150961e-06,
-      "loss": 0.9484,
+      "loss": 0.9482,
       "step": 675
     },
     {
       "epoch": 0.62,
       "learning_rate": 7.4504966644264775e-06,
-      "loss": 0.9484,
+      "loss": 0.9482,
       "step": 680
     },
     {
       "epoch": 0.63,
       "learning_rate": 7.295999523884921e-06,
-      "loss": 0.9426,
+      "loss": 0.9424,
       "step": 685
     },
     {
       "epoch": 0.63,
       "learning_rate": 7.142195648261747e-06,
-      "loss": 0.9647,
+      "loss": 0.9646,
       "step": 690
     },
     {
       "epoch": 0.64,
       "learning_rate": 6.989124470549746e-06,
-      "loss": 0.9601,
+      "loss": 0.9602,
       "step": 695
     },
     {
       "epoch": 0.64,
       "learning_rate": 6.83682523588902e-06,
-      "loss": 0.9439,
+      "loss": 0.9436,
       "step": 700
     },
     {
       "epoch": 0.65,
       "learning_rate": 6.685336991505122e-06,
-      "loss": 0.9574,
+      "loss": 0.9576,
       "step": 705
     },
     {
       "epoch": 0.65,
       "learning_rate": 6.5346985766979384e-06,
-      "loss": 0.9498,
+      "loss": 0.9497,
       "step": 710
     },
     {
       "epoch": 0.66,
       "learning_rate": 6.384948612883872e-06,
-      "loss": 0.9475,
+      "loss": 0.9474,
       "step": 715
     },
     {
       "epoch": 0.66,
       "learning_rate": 6.2361254936939e-06,
-      "loss": 0.9385,
+      "loss": 0.9382,
       "step": 720
     },
     {
       "epoch": 0.67,
       "learning_rate": 6.0882673751300235e-06,
-      "loss": 0.9499,
+      "loss": 0.9501,
       "step": 725
     },
     {
       "epoch": 0.67,
       "learning_rate": 5.941412165782645e-06,
-      "loss": 0.9487,
+      "loss": 0.9485,
       "step": 730
     },
     {
       "epoch": 0.67,
       "learning_rate": 5.79559751711138e-06,
-      "loss": 0.935,
+      "loss": 0.9349,
       "step": 735
     },
     {
       "epoch": 0.68,
       "learning_rate": 5.650860813791786e-06,
-      "loss": 0.9442,
+      "loss": 0.9444,
       "step": 740
     },
     {
       "epoch": 0.68,
       "learning_rate": 5.507239164130501e-06,
-      "loss": 0.9523,
+      "loss": 0.9522,
       "step": 745
     },
     {
       "epoch": 0.69,
       "learning_rate": 5.364769390551225e-06,
-      "loss": 0.9509,
+      "loss": 0.9507,
       "step": 750
     },
     {
@@ -923,13 +923,13 @@
     {
       "epoch": 0.7,
       "learning_rate": 5.083431275350312e-06,
-      "loss": 0.9395,
+      "loss": 0.9396,
       "step": 760
     },
     {
       "epoch": 0.7,
       "learning_rate": 4.9446350645759885e-06,
-      "loss": 0.9466,
+      "loss": 0.9467,
       "step": 765
     },
     {
@@ -941,73 +941,73 @@
     {
       "epoch": 0.71,
       "learning_rate": 4.670966253826027e-06,
-      "loss": 0.9501,
+      "loss": 0.9515,
       "step": 775
     },
     {
       "epoch": 0.72,
       "learning_rate": 4.53616381840377e-06,
-      "loss": 0.9334,
+      "loss": 0.9336,
       "step": 780
     },
     {
       "epoch": 0.72,
       "learning_rate": 4.402762228128531e-06,
-      "loss": 0.9407,
+      "loss": 0.9412,
       "step": 785
     },
     {
       "epoch": 0.72,
       "learning_rate": 4.270795685155001e-06,
-      "loss": 0.9482,
+      "loss": 0.9489,
       "step": 790
     },
     {
       "epoch": 0.73,
       "learning_rate": 4.140298023713416e-06,
-      "loss": 0.9287,
+      "loss": 0.9292,
       "step": 795
     },
     {
       "epoch": 0.73,
       "learning_rate": 4.0113027014349374e-06,
-      "loss": 0.9438,
+      "loss": 0.944,
       "step": 800
     },
     {
       "epoch": 0.74,
       "learning_rate": 3.883842790773647e-06,
-      "loss": 0.9324,
+      "loss": 0.9325,
       "step": 805
     },
     {
       "epoch": 0.74,
       "learning_rate": 3.757950970527249e-06,
-      "loss": 0.9364,
+      "loss": 0.9365,
       "step": 810
     },
     {
       "epoch": 0.75,
       "learning_rate": 3.633659517458736e-06,
-      "loss": 0.9309,
+      "loss": 0.9312,
       "step": 815
     },
     {
       "epoch": 0.75,
       "learning_rate": 3.511000298021098e-06,
-      "loss": 0.9494,
+      "loss": 0.95,
       "step": 820
     },
     {
       "epoch": 0.76,
       "learning_rate": 3.39000476018726e-06,
-      "loss": 0.9439,
+      "loss": 0.944,
       "step": 825
     },
     {
       "epoch": 0.76,
       "learning_rate": 3.2707039253872796e-06,
-      "loss": 0.9351,
+      "loss": 0.9354,
       "step": 830
     },
     {
@@ -1019,19 +1019,19 @@
     {
       "epoch": 0.77,
       "learning_rate": 3.037308270285709e-06,
-      "loss": 0.9202,
+      "loss": 0.9204,
       "step": 840
     },
     {
       "epoch": 0.78,
       "learning_rate": 2.923273289108115e-06,
-      "loss": 0.9461,
+      "loss": 0.9462,
       "step": 845
     },
     {
       "epoch": 0.78,
       "learning_rate": 2.8110526738705345e-06,
-      "loss": 0.9368,
+      "loss": 0.937,
       "step": 850
     },
     {
@@ -1043,43 +1043,43 @@
     {
       "epoch": 0.79,
       "learning_rate": 2.592169155352031e-06,
-      "loss": 0.9155,
+      "loss": 0.9157,
       "step": 860
     },
     {
       "epoch": 0.79,
       "learning_rate": 2.485562370502279e-06,
-      "loss": 0.928,
+      "loss": 0.9283,
       "step": 865
     },
     {
       "epoch": 0.8,
       "learning_rate": 2.3808821740669608e-06,
-      "loss": 0.9295,
+      "loss": 0.9294,
       "step": 870
     },
     {
       "epoch": 0.8,
       "learning_rate": 2.2781554044688015e-06,
-      "loss": 0.9402,
+      "loss": 0.9404,
       "step": 875
     },
     {
       "epoch": 0.81,
       "learning_rate": 2.1774083993013715e-06,
-      "loss": 0.9361,
+      "loss": 0.9363,
       "step": 880
     },
     {
       "epoch": 0.81,
       "learning_rate": 2.0786669885765044e-06,
-      "loss": 0.9344,
+      "loss": 0.9345,
       "step": 885
     },
     {
       "epoch": 0.82,
       "learning_rate": 1.981956488101898e-06,
-      "loss": 0.9245,
+      "loss": 0.9244,
       "step": 890
     },
     {
@@ -1091,31 +1091,31 @@
     {
       "epoch": 0.83,
       "learning_rate": 1.7947268713034128e-06,
-      "loss": 0.9483,
+      "loss": 0.9481,
       "step": 900
     },
     {
       "epoch": 0.83,
       "learning_rate": 1.704255757827963e-06,
-      "loss": 0.9351,
+      "loss": 0.935,
       "step": 905
     },
     {
       "epoch": 0.83,
       "learning_rate": 1.6159115479924259e-06,
-      "loss": 0.9237,
+      "loss": 0.9239,
       "step": 910
     },
     {
       "epoch": 0.84,
       "learning_rate": 1.529716891919074e-06,
-      "loss": 0.9211,
+      "loss": 0.9213,
       "step": 915
     },
     {
       "epoch": 0.84,
       "learning_rate": 1.4456938886170413e-06,
-      "loss": 0.938,
+      "loss": 0.9381,
       "step": 920
     },
     {
@@ -1127,31 +1127,31 @@
     {
       "epoch": 0.85,
       "learning_rate": 1.2842484469453365e-06,
-      "loss": 0.9157,
+      "loss": 0.9158,
       "step": 930
     },
     {
       "epoch": 0.86,
       "learning_rate": 1.2068674007506787e-06,
-      "loss": 0.9286,
+      "loss": 0.9288,
       "step": 935
     },
     {
       "epoch": 0.86,
       "learning_rate": 1.1317407810650372e-06,
-      "loss": 0.9214,
+      "loss": 0.9213,
       "step": 940
     },
     {
       "epoch": 0.87,
       "learning_rate": 1.0588878492200261e-06,
-      "loss": 0.9343,
+      "loss": 0.9344,
       "step": 945
     },
     {
       "epoch": 0.87,
       "learning_rate": 9.883272836080116e-07,
-      "loss": 0.9292,
+      "loss": 0.9293,
       "step": 950
     },
     {
@@ -1163,43 +1163,43 @@
     {
       "epoch": 0.88,
       "learning_rate": 8.541550213737171e-07,
-      "loss": 0.9241,
+      "loss": 0.924,
       "step": 960
     },
     {
       "epoch": 0.89,
       "learning_rate": 7.905777244947954e-07,
-      "loss": 0.9193,
+      "loss": 0.9192,
       "step": 965
     },
     {
       "epoch": 0.89,
       "learning_rate": 7.293615845160196e-07,
-      "loss": 0.9124,
+      "loss": 0.9125,
       "step": 970
     },
     {
       "epoch": 0.89,
       "learning_rate": 6.705222963319191e-07,
-      "loss": 0.928,
+      "loss": 0.9282,
       "step": 975
     },
     {
       "epoch": 0.9,
       "learning_rate": 6.140749454480932e-07,
-      "loss": 0.9382,
+      "loss": 0.9384,
       "step": 980
     },
     {
       "epoch": 0.9,
       "learning_rate": 5.600340041135133e-07,
-      "loss": 0.9258,
+      "loss": 0.9259,
       "step": 985
     },
     {
       "epoch": 0.91,
       "learning_rate": 5.0841332761005e-07,
-      "loss": 0.9275,
+      "loss": 0.9277,
       "step": 990
     },
     {
@@ -1211,7 +1211,7 @@
     {
       "epoch": 0.92,
       "learning_rate": 4.124850842338779e-07,
-      "loss": 0.925,
+      "loss": 0.9252,
       "step": 1000
     },
     {
@@ -1229,7 +1229,7 @@
     {
       "epoch": 0.93,
       "learning_rate": 2.870552305351382e-07,
-      "loss": 0.9289,
+      "loss": 0.9291,
       "step": 1015
     },
     {
@@ -1241,49 +1241,49 @@
     {
       "epoch": 0.94,
       "learning_rate": 2.158687205694443e-07,
-      "loss": 0.9214,
+      "loss": 0.9213,
       "step": 1025
     },
     {
       "epoch": 0.94,
       "learning_rate": 1.840338184455881e-07,
-      "loss": 0.9144,
+      "loss": 0.9148,
       "step": 1030
     },
     {
       "epoch": 0.95,
       "learning_rate": 1.5471558192656776e-07,
-      "loss": 0.9229,
+      "loss": 0.923,
       "step": 1035
     },
     {
       "epoch": 0.95,
       "learning_rate": 1.279215277658097e-07,
-      "loss": 0.9121,
+      "loss": 0.9124,
       "step": 1040
     },
     {
       "epoch": 0.96,
       "learning_rate": 1.0365852555447642e-07,
-      "loss": 0.9406,
+      "loss": 0.9407,
       "step": 1045
     },
     {
       "epoch": 0.96,
       "learning_rate": 8.19327959602012e-08,
-      "loss": 0.9348,
+      "loss": 0.9346,
       "step": 1050
     },
     {
       "epoch": 0.97,
       "learning_rate": 6.274990913221035e-08,
-      "loss": 0.9201,
+      "loss": 0.9198,
       "step": 1055
     },
     {
       "epoch": 0.97,
       "learning_rate": 4.6114783273213395e-08,
-      "loss": 0.9212,
+      "loss": 0.9214,
       "step": 1060
     },
     {
@@ -1295,13 +1295,13 @@
     {
       "epoch": 0.98,
       "learning_rate": 2.05042201422323e-08,
-      "loss": 0.9265,
+      "loss": 0.9268,
       "step": 1070
     },
     {
       "epoch": 0.99,
       "learning_rate": 1.1535349032167908e-08,
-      "loss": 0.9244,
+      "loss": 0.9249,
       "step": 1075
     },
     {
@@ -1313,31 +1313,31 @@
     {
       "epoch": 1.0,
       "learning_rate": 1.2819245493955746e-09,
-      "loss": 0.916,
+      "loss": 0.9161,
       "step": 1085
     },
     {
       "epoch": 1.0,
       "learning_rate": 0.0,
-      "loss": 0.9078,
+      "loss": 0.9077,
       "step": 1090
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.9352946281433105,
-      "eval_runtime": 143.3511,
-      "eval_samples_per_second": 107.645,
-      "eval_steps_per_second": 1.688,
+      "eval_loss": 0.9353834986686707,
+      "eval_runtime": 144.309,
+      "eval_samples_per_second": 106.93,
+      "eval_steps_per_second": 1.677,
       "step": 1090
     },
     {
       "epoch": 1.0,
       "step": 1090,
       "total_flos": 456447649382400.0,
-      "train_loss": 0.07620820999145508,
-      "train_runtime": 563.6077,
-      "train_samples_per_second": 247.381,
-      "train_steps_per_second": 1.934
+      "train_loss": 0.9776670060026537,
+      "train_runtime": 5614.3798,
+      "train_samples_per_second": 24.834,
+      "train_steps_per_second": 0.194
     }
   ],
   "logging_steps": 5,