Upload 15 files

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83cc74d0d5909d2d8879aa922fb58039963c6f3cd976de52605d9d19608a23ff
+oid sha256:470ad931266a6e96ba5a885406ab42d02b88b72338581c4a86550f9cc7d6ca17
 size 2384234968

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6fcac12a7c6675631610843bec77e53d42633ba080c920923a68c4636448ada
+oid sha256:cde182865a6d3d24b73e9485347b908e0449d66bb6e9edccc931a636295d2561
 size 4768667262

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2b5691046896d865f067a1958689168fc2411c74d2f82d596bd6a636b2b141b
+oid sha256:9196a1e708bf24d6abba41cce3f8558820acc3e50f9394c5955e29eb41ffea3d
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80938d2c342a9e9feb870ab78d4bc6c71c45abde7ade4df8bd1b2b97febc4209
+oid sha256:e9f54374167c346852156307ccec6d5c312ed574fd353a754e476da6abc305ce
 size 1064

trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.30208,
-  "eval_steps": 100,
-  "global_step": 900,
+  "epoch": 1.79104,
+  "eval_steps": 700,
+  "global_step": 700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -16,147 +16,118 @@
       "loss": 1.0238,
       "step": 1
     },
+    {
+      "epoch": 0.128,
+      "grad_norm": 3.0013201236724854,
+      "learning_rate": 2.0851063829787235e-06,
+      "loss": 0.8222,
+      "step": 50
+    },
     {
       "epoch": 0.256,
       "grad_norm": 2.3984222412109375,
       "learning_rate": 4.212765957446809e-06,
-      "loss": 0.6317,
+      "loss": 0.445,
       "step": 100
     },
     {
-      "epoch": 0.256,
-      "eval_loss": 0.3901349604129791,
-      "eval_runtime": 76.1829,
-      "eval_samples_per_second": 16.408,
-      "eval_steps_per_second": 8.204,
-      "step": 100
+      "epoch": 0.384,
+      "grad_norm": 2.368490695953369,
+      "learning_rate": 6.3404255319148945e-06,
+      "loss": 0.4149,
+      "step": 150
     },
     {
       "epoch": 0.512,
       "grad_norm": 2.158310890197754,
       "learning_rate": 8.46808510638298e-06,
-      "loss": 0.4072,
+      "loss": 0.3995,
       "step": 200
     },
     {
-      "epoch": 0.512,
-      "eval_loss": 0.3636568784713745,
-      "eval_runtime": 76.1675,
-      "eval_samples_per_second": 16.411,
-      "eval_steps_per_second": 8.206,
-      "step": 200
+      "epoch": 0.64,
+      "grad_norm": 2.212986707687378,
+      "learning_rate": 9.994504457428557e-06,
+      "loss": 0.3701,
+      "step": 250
     },
     {
       "epoch": 0.768,
-      "grad_norm": 2.0863559246063232,
+      "grad_norm": 2.0821216106414795,
       "learning_rate": 9.885572470188207e-06,
-      "loss": 0.3826,
+      "loss": 0.3951,
       "step": 300
     },
     {
-      "epoch": 0.768,
-      "eval_loss": 0.3526187837123871,
-      "eval_runtime": 76.2073,
-      "eval_samples_per_second": 16.403,
-      "eval_steps_per_second": 8.201,
-      "step": 300
+      "epoch": 0.896,
+      "grad_norm": 2.1344685554504395,
+      "learning_rate": 9.639951230825433e-06,
+      "loss": 0.3777,
+      "step": 350
     },
     {
       "epoch": 1.02304,
-      "grad_norm": 2.2784180641174316,
+      "grad_norm": 2.2334532737731934,
       "learning_rate": 9.2645127658466e-06,
-      "loss": 0.3653,
+      "loss": 0.353,
       "step": 400
     },
     {
-      "epoch": 1.02304,
-      "eval_loss": 0.3516922891139984,
-      "eval_runtime": 76.1603,
-      "eval_samples_per_second": 16.413,
-      "eval_steps_per_second": 8.206,
-      "step": 400
+      "epoch": 1.15104,
+      "grad_norm": 2.0466365814208984,
+      "learning_rate": 8.76976114684395e-06,
+      "loss": 0.3095,
+      "step": 450
     },
     {
       "epoch": 1.27904,
-      "grad_norm": 2.1337344646453857,
+      "grad_norm": 2.1597046852111816,
       "learning_rate": 8.169538606063647e-06,
-      "loss": 0.3122,
+      "loss": 0.3148,
       "step": 500
     },
     {
-      "epoch": 1.27904,
-      "eval_loss": 0.3464358150959015,
-      "eval_runtime": 76.195,
-      "eval_samples_per_second": 16.405,
-      "eval_steps_per_second": 8.203,
-      "step": 500
+      "epoch": 1.40704,
+      "grad_norm": 2.369905471801758,
+      "learning_rate": 7.480638256432977e-06,
+      "loss": 0.3089,
+      "step": 550
     },
     {
       "epoch": 1.53504,
-      "grad_norm": 2.1311352252960205,
+      "grad_norm": 2.1879630088806152,
       "learning_rate": 6.722334251421665e-06,
-      "loss": 0.3107,
+      "loss": 0.3123,
       "step": 600
     },
     {
-      "epoch": 1.53504,
-      "eval_loss": 0.34351983666419983,
-      "eval_runtime": 76.1475,
-      "eval_samples_per_second": 16.415,
-      "eval_steps_per_second": 8.208,
-      "step": 600
+      "epoch": 1.66304,
+      "grad_norm": 1.9208930730819702,
+      "learning_rate": 5.915842529992632e-06,
+      "loss": 0.2917,
+      "step": 650
     },
     {
       "epoch": 1.79104,
-      "grad_norm": 1.905997633934021,
+      "grad_norm": 1.9313366413116455,
       "learning_rate": 5.083727233997775e-06,
-      "loss": 0.301,
+      "loss": 0.3105,
       "step": 700
     },
     {
       "epoch": 1.79104,
-      "eval_loss": 0.3379676640033722,
-      "eval_runtime": 76.2095,
-      "eval_samples_per_second": 16.402,
-      "eval_steps_per_second": 8.201,
+      "eval_loss": 0.3375703692436218,
+      "eval_runtime": 78.8007,
+      "eval_samples_per_second": 15.863,
+      "eval_steps_per_second": 7.931,
       "step": 700
-    },
-    {
-      "epoch": 2.04608,
-      "grad_norm": 2.4192748069763184,
-      "learning_rate": 3.4358156257133644e-06,
-      "loss": 0.2912,
-      "step": 800
-    },
-    {
-      "epoch": 2.04608,
-      "eval_loss": 0.35626986622810364,
-      "eval_runtime": 76.1989,
-      "eval_samples_per_second": 16.404,
-      "eval_steps_per_second": 8.202,
-      "step": 800
-    },
-    {
-      "epoch": 2.30208,
-      "grad_norm": 2.3621625900268555,
-      "learning_rate": 1.9617315158214363e-06,
-      "loss": 0.2201,
-      "step": 900
-    },
-    {
-      "epoch": 2.30208,
-      "eval_loss": 0.3607979118824005,
-      "eval_runtime": 76.2076,
-      "eval_samples_per_second": 16.403,
-      "eval_steps_per_second": 8.201,
-      "step": 900
     }
   ],
-  "logging_steps": 100,
+  "logging_steps": 50,
   "max_steps": 1173,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
-  "save_steps": 900,
+  "save_steps": 700,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {
@@ -169,7 +140,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.5574906692226253e+17,
+  "total_flos": 1.2117424625575526e+17,
   "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:c7ee02a0cd4f7b137ba78feabcbed64937cec52a6b8a3f66a9f57da227b8f821
+oid sha256:4dc9cd2c46e7e2983902b8d6f466821351988fe7f14f96e63e9e4fa6d81ad500
 size 5304