Training in progress, step 100, checkpoint

last-checkpoint/README.md

@@ -236,4 +236,23 @@ The following `bitsandbytes` quantization config was used during training:
 ### Framework versions
 
 
+- PEFT 0.6.3.dev0
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- load_in_8bit: True
+- load_in_4bit: False
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: fp4
+- bnb_4bit_use_double_quant: False
+- bnb_4bit_compute_dtype: float32
+
+### Framework versions
+
+
 - PEFT 0.6.3.dev0

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef2c8347f1be9aebfc1271f5f427065d90f0e50efb3a58333837af0c92cfabd6
+oid sha256:a48e0e5bde33bb82c52da7473d5f0b81aa3ad4d16f1b412eac2591e9ab1a27e1
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2323219fe41e5b63b7d472b1ca12c211f60c05cc1bc4517c5ee44eb63d4d78ba
+oid sha256:85ad78154fee72f2aca6fca0e93077aac0cc3d7717cd9c9ea57c8c724ce6383b
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbebb324d09ed3b79588ce704d8e867cae0d7b652313decf15a03bfb1ba6792e
+oid sha256:43b3279df66a37918fc86e90e9c6a8fa486ff8c675a0a5adbdb77a496e973bf1
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5efc416a6883409dd7ab6f5c779e107c7c2baa7af6e12ed9fbd9dd73b8b20784
+oid sha256:216f76b8039f833c337db298c81f13b12082d5fd4f9d866cecd34b2ca7550b37
 size 627

last-checkpoint/tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:490b3432398e9c45dda0581bf34f6daa9411bda59cd0cd831670d7c86f90d2dd
-size 14500570
+oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
+size 14500471

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.005772113800049,
-  "best_model_checkpoint": "./outputs/checkpoint-500",
-  "epoch": 0.7102272727272727,
+  "best_metric": 2.11466121673584,
+  "best_model_checkpoint": "./outputs/checkpoint-100",
+  "epoch": 0.14204545454545456,
   "eval_steps": 100,
-  "global_step": 500,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -11,72 +11,16 @@
     {
       "epoch": 0.14,
       "learning_rate": 0.0002,
-      "loss": 2.4473,
+      "loss": 2.4493,
       "step": 100
     },
     {
       "epoch": 0.14,
-      "eval_loss": 2.1156835556030273,
-      "eval_runtime": 56.5007,
-      "eval_samples_per_second": 26.407,
-      "eval_steps_per_second": 3.31,
+      "eval_loss": 2.11466121673584,
+      "eval_runtime": 56.3914,
+      "eval_samples_per_second": 26.458,
+      "eval_steps_per_second": 3.316,
       "step": 100
-    },
-    {
-      "epoch": 0.28,
-      "learning_rate": 0.0002,
-      "loss": 2.253,
-      "step": 200
-    },
-    {
-      "epoch": 0.28,
-      "eval_loss": 2.073469877243042,
-      "eval_runtime": 54.097,
-      "eval_samples_per_second": 27.58,
-      "eval_steps_per_second": 3.457,
-      "step": 200
-    },
-    {
-      "epoch": 0.43,
-      "learning_rate": 0.0002,
-      "loss": 2.2236,
-      "step": 300
-    },
-    {
-      "epoch": 0.43,
-      "eval_loss": 2.046644687652588,
-      "eval_runtime": 53.9984,
-      "eval_samples_per_second": 27.63,
-      "eval_steps_per_second": 3.463,
-      "step": 300
-    },
-    {
-      "epoch": 0.57,
-      "learning_rate": 0.0002,
-      "loss": 2.1911,
-      "step": 400
-    },
-    {
-      "epoch": 0.57,
-      "eval_loss": 2.0298001766204834,
-      "eval_runtime": 54.0695,
-      "eval_samples_per_second": 27.594,
-      "eval_steps_per_second": 3.459,
-      "step": 400
-    },
-    {
-      "epoch": 0.71,
-      "learning_rate": 0.0002,
-      "loss": 2.1704,
-      "step": 500
-    },
-    {
-      "epoch": 0.71,
-      "eval_loss": 2.005772113800049,
-      "eval_runtime": 54.1197,
-      "eval_samples_per_second": 27.569,
-      "eval_steps_per_second": 3.455,
-      "step": 500
     }
   ],
   "logging_steps": 100,
@@ -84,7 +28,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.52069905096704e+16,
+  "total_flos": 3017571420487680.0,
   "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:f41f795a4d38cf0b20c4ab7c8b00a3bbdbb028ecd4a46fa726fb31c2bc81f568
+oid sha256:127332f99b100bbb3a9659e58e9fcfa7f341ada72cf3863477eab4b88b505759
 size 4283