Training in progress, step 1900, 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:a63c5195f49dc20073b6b26d804dea6c2185be3ac0d662446cdfd43a0952aa27
+oid sha256:33586be5fbd82fae9ac6620008892678b5f2d7dd7e5a8ef6a3da78d07653c26b
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3dd44b1006cd7a5021c7b278b6f9343ab977242abc34234321f4d0a1dbecd537
-size 100693001
+oid sha256:12cd486f70d22320e3dea3f1a140dee210d82739ecaf04b0b58ec946113e98af
+size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b671a8eca01cb9b6f4736829516f37a248a3b5268cb765ba18a2dd729fd3c17b
+oid sha256:0d9a5e96d5319f578e3676c8a6aaa1d019c2f577ff8abb55dcf5d74b47d9273c
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e2a7616f99506ce2115c2b51a0c8d6af99dd5ac2b72ef98b701eba7ab661118
+oid sha256:b2d4b4cadcabd3e17adcc39c43d8312038a1f5c7f53125f288ba4de5ed25396b
 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": 1.4274669885635376,
-  "best_model_checkpoint": "./outputs/checkpoint-2200",
-  "epoch": 1.6029143897996359,
+  "best_metric": 1.4694663286209106,
+  "best_model_checkpoint": "./outputs/checkpoint-1900",
+  "epoch": 1.384335154826958,
   "eval_steps": 100,
-  "global_step": 2200,
+  "global_step": 1900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -151,177 +151,135 @@
     {
       "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,
-      "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,
-      "step": 2000
-    },
-    {
-      "epoch": 1.53,
-      "learning_rate": 0.0002,
-      "loss": 1.4358,
-      "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,
-      "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
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.3116148866861466e+17,
+  "total_flos": 1.1327940508889088e+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