Training in progress, step 2100, checkpoint

last-checkpoint/README.md

@@ -216,4 +216,11 @@ The following `bitsandbytes` quantization config was used during training:
 ### Framework versions
 
 
+- PEFT 0.6.0.dev0
+## Training procedure
+
+
+### Framework versions
+
+
 - PEFT 0.6.0.dev0

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6269c590138d0bc20110051e91990b7b4bab17315b95ec4a72df38ec0d08107
+oid sha256:c103417eaf6818fff52453461d8de8c3bb3b816b606af4dc62e6f66e3d0d1421
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f01636dc6b615cd115ecabea1ebae15d90eb4974df2f3205e463f01906338321
-size 100693001
+oid sha256:6e48e874b3471f5c826f024130566f3cfc276e31949eff0dd2a1b999b19e9c41
+size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91ce772f9b2a32aa1626e6ccbc135efad31ff1869adac26764a35a5c73bb1216
+oid sha256:2eac30a54a346eb77c88df167a4adbf13ea272bd1dd65fd9f0292631c36fab66
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da625bdbaba1e1b9a41c85e65d17da2b21a48f48126a4ef7c6978303b33ffaa8
+oid sha256:b1238b3bfa75e49a19396161e9e7b72ab89cdd1a3f63b51c0ab4d6e8d216c5a5
 size 627

last-checkpoint/special_tokens_map.json

@@ -1,6 +1,24 @@
 {
-  "bos_token": "<|endoftext|>",
-  "eos_token": "<|endoftext|>",
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "pad_token": "<|endoftext|>",
-  "unk_token": "<|endoftext|>"
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
 }

last-checkpoint/tokenizer_config.json

@@ -13,8 +13,12 @@
   "bos_token": "<|endoftext|>",
   "clean_up_tokenization_spaces": true,
   "eos_token": "<|endoftext|>",
+  "max_length": 512,
   "model_max_length": 1024,
   "pad_token": "<|endoftext|>",
+  "stride": 0,
   "tokenizer_class": "GPT2Tokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
   "unk_token": "<|endoftext|>"
 }

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.4145855903625488,
-  "best_model_checkpoint": "./outputs/checkpoint-2300",
-  "epoch": 1.6757741347905282,
+  "best_metric": 1.4383305311203003,
+  "best_model_checkpoint": "./outputs/checkpoint-2100",
+  "epoch": 1.530054644808743,
   "eval_steps": 100,
-  "global_step": 2300,
+  "global_step": 2100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -151,191 +151,163 @@
     {
       "epoch": 0.8,
       "learning_rate": 0.0002,
-      "loss": 1.6268,
+      "loss": 1.6461,
       "step": 1100
     },
     {
       "epoch": 0.8,
-      "eval_loss": 1.6125303506851196,
-      "eval_runtime": 144.0648,
-      "eval_samples_per_second": 43.55,
-      "eval_steps_per_second": 5.449,
+      "eval_loss": 1.6240431070327759,
+      "eval_runtime": 293.5989,
+      "eval_samples_per_second": 21.369,
+      "eval_steps_per_second": 2.674,
       "step": 1100
     },
     {
       "epoch": 0.87,
       "learning_rate": 0.0002,
-      "loss": 1.5911,
+      "loss": 1.5992,
       "step": 1200
     },
     {
       "epoch": 0.87,
-      "eval_loss": 1.5925209522247314,
-      "eval_runtime": 144.042,
-      "eval_samples_per_second": 43.557,
-      "eval_steps_per_second": 5.45,
+      "eval_loss": 1.5974311828613281,
+      "eval_runtime": 291.7,
+      "eval_samples_per_second": 21.508,
+      "eval_steps_per_second": 2.691,
       "step": 1200
     },
     {
       "epoch": 0.95,
       "learning_rate": 0.0002,
-      "loss": 1.5986,
+      "loss": 1.6021,
       "step": 1300
     },
     {
       "epoch": 0.95,
-      "eval_loss": 1.571681261062622,
-      "eval_runtime": 144.1746,
-      "eval_samples_per_second": 43.517,
-      "eval_steps_per_second": 5.445,
+      "eval_loss": 1.5751127004623413,
+      "eval_runtime": 289.9524,
+      "eval_samples_per_second": 21.638,
+      "eval_steps_per_second": 2.707,
       "step": 1300
     },
     {
       "epoch": 1.02,
       "learning_rate": 0.0002,
-      "loss": 1.5514,
+      "loss": 1.5538,
       "step": 1400
     },
     {
       "epoch": 1.02,
-      "eval_loss": 1.5524405241012573,
-      "eval_runtime": 183.3697,
-      "eval_samples_per_second": 34.215,
-      "eval_steps_per_second": 4.281,
+      "eval_loss": 1.5539450645446777,
+      "eval_runtime": 287.8748,
+      "eval_samples_per_second": 21.794,
+      "eval_steps_per_second": 2.727,
       "step": 1400
     },
     {
       "epoch": 1.09,
       "learning_rate": 0.0002,
-      "loss": 1.5235,
+      "loss": 1.5249,
       "step": 1500
     },
     {
       "epoch": 1.09,
-      "eval_loss": 1.534788966178894,
-      "eval_runtime": 317.3733,
-      "eval_samples_per_second": 19.769,
-      "eval_steps_per_second": 2.473,
+      "eval_loss": 1.5348094701766968,
+      "eval_runtime": 287.891,
+      "eval_samples_per_second": 21.793,
+      "eval_steps_per_second": 2.727,
       "step": 1500
     },
     {
       "epoch": 1.17,
       "learning_rate": 0.0002,
-      "loss": 1.5051,
+      "loss": 1.506,
       "step": 1600
     },
     {
       "epoch": 1.17,
-      "eval_loss": 1.517040491104126,
-      "eval_runtime": 315.5897,
-      "eval_samples_per_second": 19.88,
-      "eval_steps_per_second": 2.487,
+      "eval_loss": 1.515953540802002,
+      "eval_runtime": 289.836,
+      "eval_samples_per_second": 21.647,
+      "eval_steps_per_second": 2.708,
       "step": 1600
     },
     {
       "epoch": 1.24,
       "learning_rate": 0.0002,
-      "loss": 1.5036,
+      "loss": 1.5042,
       "step": 1700
     },
     {
       "epoch": 1.24,
-      "eval_loss": 1.500235915184021,
-      "eval_runtime": 314.1201,
-      "eval_samples_per_second": 19.973,
-      "eval_steps_per_second": 2.499,
+      "eval_loss": 1.4988901615142822,
+      "eval_runtime": 291.5471,
+      "eval_samples_per_second": 21.52,
+      "eval_steps_per_second": 2.693,
       "step": 1700
     },
     {
       "epoch": 1.31,
       "learning_rate": 0.0002,
-      "loss": 1.4767,
+      "loss": 1.4762,
       "step": 1800
     },
     {
       "epoch": 1.31,
-      "eval_loss": 1.4854458570480347,
-      "eval_runtime": 313.3904,
-      "eval_samples_per_second": 20.02,
-      "eval_steps_per_second": 2.505,
+      "eval_loss": 1.4844294786453247,
+      "eval_runtime": 293.6668,
+      "eval_samples_per_second": 21.364,
+      "eval_steps_per_second": 2.673,
       "step": 1800
     },
     {
       "epoch": 1.38,
       "learning_rate": 0.0002,
-      "loss": 1.4665,
+      "loss": 1.4652,
       "step": 1900
     },
     {
       "epoch": 1.38,
-      "eval_loss": 1.4697930812835693,
-      "eval_runtime": 314.584,
-      "eval_samples_per_second": 19.944,
-      "eval_steps_per_second": 2.495,
+      "eval_loss": 1.4694663286209106,
+      "eval_runtime": 295.5867,
+      "eval_samples_per_second": 21.226,
+      "eval_steps_per_second": 2.656,
       "step": 1900
     },
     {
       "epoch": 1.46,
       "learning_rate": 0.0002,
-      "loss": 1.4498,
+      "loss": 1.4481,
       "step": 2000
     },
     {
       "epoch": 1.46,
-      "eval_loss": 1.456108808517456,
-      "eval_runtime": 316.2748,
-      "eval_samples_per_second": 19.837,
-      "eval_steps_per_second": 2.482,
+      "eval_loss": 1.4534634351730347,
+      "eval_runtime": 296.7451,
+      "eval_samples_per_second": 21.143,
+      "eval_steps_per_second": 2.645,
       "step": 2000
     },
     {
       "epoch": 1.53,
       "learning_rate": 0.0002,
-      "loss": 1.4358,
+      "loss": 1.4335,
       "step": 2100
     },
     {
       "epoch": 1.53,
-      "eval_loss": 1.4408069849014282,
-      "eval_runtime": 317.862,
-      "eval_samples_per_second": 19.738,
-      "eval_steps_per_second": 2.47,
+      "eval_loss": 1.4383305311203003,
+      "eval_runtime": 294.659,
+      "eval_samples_per_second": 21.292,
+      "eval_steps_per_second": 2.664,
       "step": 2100
-    },
-    {
-      "epoch": 1.6,
-      "learning_rate": 0.0002,
-      "loss": 1.4099,
-      "step": 2200
-    },
-    {
-      "epoch": 1.6,
-      "eval_loss": 1.4274669885635376,
-      "eval_runtime": 319.6918,
-      "eval_samples_per_second": 19.625,
-      "eval_steps_per_second": 2.455,
-      "step": 2200
-    },
-    {
-      "epoch": 1.68,
-      "learning_rate": 0.0002,
-      "loss": 1.409,
-      "step": 2300
-    },
-    {
-      "epoch": 1.68,
-      "eval_loss": 1.4145855903625488,
-      "eval_runtime": 320.1441,
-      "eval_samples_per_second": 19.597,
-      "eval_steps_per_second": 2.452,
-      "step": 2300
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.3711720075820237e+17,
+  "total_flos": 1.2520111653502157e+17,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:435ef416a520c327b44f1a335ae059bdb8b9a978d39dfecd5bff01684de2670c
+oid sha256:e1ab4a714067324690a64db56d021644d50462360424c147b3e9df6b69650fa0
 size 4155