Training in progress, step 50, checkpoint

checkpoint-50/README.md

@@ -216,23 +216,4 @@ The following `bitsandbytes` quantization config was used during training:
 ### Framework versions
 
 
-- PEFT 0.6.0
-## 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 0.6.0

checkpoint-50/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4166664b368ce4cd6e33f9c9e5988f4cd0f67acb476bfaf70382c8384056788a
+oid sha256:6b18bddea3393e2c36e7792c11a181c1109d4ad2e0dc3c12f2a473b7c3b1bda9
 size 19690328

checkpoint-50/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b45841b33a84da2a1b486d9551765151b191da5b910c16259aacd2114b08f224
+oid sha256:683100401d1d170b7e951c495ee126773e47c993a54b3e4533f95a562aa15deb
 size 38087098

checkpoint-50/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f197d949afeca5f73d2f9c4e8008213842dfb0a96b35f9276fd815dc3b955de7
+oid sha256:e8a9d78310659bc91e8466ed6f6b24b5d5c93728931c321499a58eecfcf00f1b
 size 14244

checkpoint-50/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:609839f04f1b0fbf1991d0fad32a3ffd669d0403ec0a02950e111710295f6725
+oid sha256:b3edbdbe5702101dc9e07140e67378112cf5e69efff3173e946a6c6a3d43c33f
 size 1064

checkpoint-50/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 6.581231117248535,
+  "best_metric": 0.6264824271202087,
   "best_model_checkpoint": "output/checkpoint-50",
-  "epoch": 0.2552648372686662,
+  "epoch": 0.040108292389451516,
   "eval_steps": 50,
   "global_step": 50,
   "is_hyper_param_search": false,
@@ -9,32 +9,32 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.01,
-      "learning_rate": 2.777777777777778e-07,
-      "loss": 9.5569,
+      "epoch": 0.0,
+      "learning_rate": 0.0,
+      "loss": 1.6453,
       "step": 1
     },
     {
-      "epoch": 0.26,
-      "learning_rate": 1.25e-05,
-      "loss": 6.958,
+      "epoch": 0.04,
+      "learning_rate": 9.777777777777779e-06,
+      "loss": 1.3216,
       "step": 50
     },
     {
-      "epoch": 0.26,
-      "eval_accuracy": 0.42857142857142855,
-      "eval_loss": 6.581231117248535,
-      "eval_runtime": 28.2966,
-      "eval_samples_per_second": 6.184,
-      "eval_steps_per_second": 1.555,
+      "epoch": 0.04,
+      "eval_accuracy": 0.8057142857142857,
+      "eval_loss": 0.6264824271202087,
+      "eval_runtime": 28.8159,
+      "eval_samples_per_second": 6.073,
+      "eval_steps_per_second": 1.527,
       "step": 50
     }
   ],
   "logging_steps": 50,
-  "max_steps": 1000,
-  "num_train_epochs": 6,
+  "max_steps": 1246,
+  "num_train_epochs": 1,
   "save_steps": 50,
-  "total_flos": 1.854149272129536e+16,
+  "total_flos": 1.843789564379136e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-50/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a0ef36d7c31265f68a3edc8ae53ce9272dcc90a79f261e29a2a46e040c77a33
+oid sha256:2b7c1677f66b49d74d914a320393e2fa29530a9670d5824a30beac9590468d16
 size 4600