Upload folder using huggingface_hub

README.md

@@ -3,20 +3,6 @@ library_name: peft
 ---
 ## Training procedure
 
-
-The following `bitsandbytes` quantization config was used during training:
-- quant_method: bitsandbytes
-- _load_in_8bit: False
-- _load_in_4bit: True
-- 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: nf4
-- bnb_4bit_use_double_quant: True
-- bnb_4bit_compute_dtype: bfloat16
-- load_in_4bit: True
-- load_in_8bit: False
 ### Framework versions
 
 

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44dca49c744beb5b5154de61adad3fa409959e53d3efe0b94d4b6b4be819c8c7
-size 160102926
+oid sha256:b1a86700c1252c736346a01d5363786e9fcfede8e32722382df82888062649dc
+size 160086542

all_results.json

@@ -1,8 +1,17 @@
 {
     "epoch": 1.0,
+    "eval_accuracy": 0.86,
+    "eval_f1_score": 0.8603998368013055,
+    "eval_gmean": 0.8596058209945328,
+    "eval_loss": 0.27683594822883606,
+    "eval_precision": 0.8614935064935065,
+    "eval_recall": 0.86,
+    "eval_runtime": 306.7277,
+    "eval_samples_per_second": 0.326,
+    "eval_steps_per_second": 0.042,
     "total_flos": 1.0505428534296576e+17,
-    "train_loss": 0.5505467119013391,
-    "train_runtime": 36020.1073,
-    "train_samples_per_second": 0.146,
-    "train_steps_per_second": 0.002
+    "train_loss": 0.366200889029154,
+    "train_runtime": 19534.4865,
+    "train_samples_per_second": 0.269,
+    "train_steps_per_second": 0.004
 }

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 1.0,
+    "eval_accuracy": 0.86,
+    "eval_f1_score": 0.8603998368013055,
+    "eval_gmean": 0.8596058209945328,
+    "eval_loss": 0.27683594822883606,
+    "eval_precision": 0.8614935064935065,
+    "eval_recall": 0.86,
+    "eval_runtime": 306.7277,
+    "eval_samples_per_second": 0.326,
+    "eval_steps_per_second": 0.042
+}

runs/Mar21_15-36-53_d6d12f13090d/events.out.tfevents.1711035415.d6d12f13090d.34.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29b8e193b83b01eb3a91c7caf4df259e599b61e5b7c371411c1c93db71fd0b04
+size 12485

runs/Mar21_15-36-53_d6d12f13090d/events.out.tfevents.1711055257.d6d12f13090d.34.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e3d5deb2bf6a997ff74edf80a28e796d26741b5b0c12acbd086cd4fb9e11e1c
+size 605

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "total_flos": 1.0505428534296576e+17,
-    "train_loss": 0.5505467119013391,
-    "train_runtime": 36020.1073,
-    "train_samples_per_second": 0.146,
-    "train_steps_per_second": 0.002
+    "train_loss": 0.366200889029154,
+    "train_runtime": 19534.4865,
+    "train_samples_per_second": 0.269,
+    "train_steps_per_second": 0.004
 }

trainer_state.json

@@ -10,212 +10,212 @@
   "log_history": [
     {
       "epoch": 0.1,
-      "grad_norm": 12.02409839630127,
+      "grad_norm": 5.78125,
       "learning_rate": 1.8500000000000002e-05,
-      "loss": 1.0056,
+      "loss": 0.3492,
       "step": 8
     },
     {
       "epoch": 0.1,
-      "eval_accuracy": 0.76,
-      "eval_f1_score": 0.76,
-      "eval_gmean": 0.7536682052115493,
-      "eval_loss": 0.6476712226867676,
-      "eval_precision": 0.76,
-      "eval_recall": 0.76,
-      "eval_runtime": 211.0576,
-      "eval_samples_per_second": 0.474,
-      "eval_steps_per_second": 0.062,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.27931639552116394,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.5917,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 8
     },
     {
       "epoch": 0.19,
-      "grad_norm": 13.794071197509766,
+      "grad_norm": 6.09375,
       "learning_rate": 1.65e-05,
-      "loss": 0.761,
+      "loss": 0.2969,
       "step": 16
     },
     {
       "epoch": 0.19,
-      "eval_accuracy": 0.8,
-      "eval_f1_score": 0.8,
-      "eval_gmean": 0.7947281139113244,
-      "eval_loss": 0.5526618361473083,
-      "eval_precision": 0.8,
-      "eval_recall": 0.8,
-      "eval_runtime": 211.7038,
-      "eval_samples_per_second": 0.472,
-      "eval_steps_per_second": 0.061,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.2782519459724426,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.7174,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 16
     },
     {
       "epoch": 0.29,
-      "grad_norm": 9.750018119812012,
+      "grad_norm": 8.125,
       "learning_rate": 1.45e-05,
-      "loss": 0.638,
+      "loss": 0.4192,
       "step": 24
     },
     {
       "epoch": 0.29,
-      "eval_accuracy": 0.81,
-      "eval_f1_score": 0.8102915601023019,
-      "eval_gmean": 0.8066432910616553,
-      "eval_loss": 0.4845515191555023,
-      "eval_precision": 0.8107466340269278,
-      "eval_recall": 0.81,
-      "eval_runtime": 211.1147,
-      "eval_samples_per_second": 0.474,
-      "eval_steps_per_second": 0.062,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.2776171863079071,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.6283,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 24
     },
     {
       "epoch": 0.39,
-      "grad_norm": 9.329747200012207,
+      "grad_norm": 7.65625,
       "learning_rate": 1.25e-05,
-      "loss": 0.6367,
+      "loss": 0.3537,
       "step": 32
     },
     {
       "epoch": 0.39,
-      "eval_accuracy": 0.81,
-      "eval_f1_score": 0.8102915601023019,
-      "eval_gmean": 0.8066432910616553,
-      "eval_loss": 0.4284454584121704,
-      "eval_precision": 0.8107466340269278,
-      "eval_recall": 0.81,
-      "eval_runtime": 210.7064,
-      "eval_samples_per_second": 0.475,
-      "eval_steps_per_second": 0.062,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.27734375,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.9153,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 32
     },
     {
       "epoch": 0.49,
-      "grad_norm": 9.867768287658691,
+      "grad_norm": 6.46875,
       "learning_rate": 1.0500000000000001e-05,
-      "loss": 0.5932,
+      "loss": 0.2969,
       "step": 40
     },
     {
       "epoch": 0.49,
-      "eval_accuracy": 0.82,
-      "eval_f1_score": 0.82,
-      "eval_gmean": 0.8152572201307815,
-      "eval_loss": 0.3858851194381714,
-      "eval_precision": 0.82,
-      "eval_recall": 0.82,
-      "eval_runtime": 211.0479,
-      "eval_samples_per_second": 0.474,
-      "eval_steps_per_second": 0.062,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.2774414122104645,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.7575,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 40
     },
     {
       "epoch": 0.58,
-      "grad_norm": 10.276468276977539,
+      "grad_norm": 7.375,
       "learning_rate": 8.5e-06,
-      "loss": 0.4603,
+      "loss": 0.3783,
       "step": 48
     },
     {
       "epoch": 0.58,
-      "eval_accuracy": 0.83,
-      "eval_f1_score": 0.8297023993409536,
-      "eval_gmean": 0.823868981405519,
-      "eval_loss": 0.3537224233150482,
-      "eval_precision": 0.8295742042166185,
-      "eval_recall": 0.83,
-      "eval_runtime": 212.1782,
-      "eval_samples_per_second": 0.471,
-      "eval_steps_per_second": 0.061,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.27705079317092896,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.7281,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 48
     },
     {
       "epoch": 0.68,
-      "grad_norm": 8.40848445892334,
+      "grad_norm": 7.09375,
       "learning_rate": 6.5000000000000004e-06,
-      "loss": 0.3794,
+      "loss": 0.4144,
       "step": 56
     },
     {
       "epoch": 0.68,
       "eval_accuracy": 0.86,
-      "eval_f1_score": 0.86,
-      "eval_gmean": 0.8563139396380225,
-      "eval_loss": 0.3282211720943451,
-      "eval_precision": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.27685546875,
+      "eval_precision": 0.8614935064935065,
       "eval_recall": 0.86,
-      "eval_runtime": 211.0187,
-      "eval_samples_per_second": 0.474,
-      "eval_steps_per_second": 0.062,
+      "eval_runtime": 306.848,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 56
     },
     {
       "epoch": 0.78,
-      "grad_norm": 8.91843032836914,
+      "grad_norm": 5.6875,
       "learning_rate": 4.5e-06,
-      "loss": 0.4213,
+      "loss": 0.4167,
       "step": 64
     },
     {
       "epoch": 0.78,
-      "eval_accuracy": 0.87,
-      "eval_f1_score": 0.8701994884910484,
-      "eval_gmean": 0.8682247898524239,
-      "eval_loss": 0.31104254722595215,
-      "eval_precision": 0.8705752753977967,
-      "eval_recall": 0.87,
-      "eval_runtime": 211.0456,
-      "eval_samples_per_second": 0.474,
-      "eval_steps_per_second": 0.062,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.2767578065395355,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.3539,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 64
     },
     {
       "epoch": 0.88,
-      "grad_norm": 6.700900554656982,
+      "grad_norm": 5.34375,
       "learning_rate": 2.5e-06,
-      "loss": 0.3611,
+      "loss": 0.3515,
       "step": 72
     },
     {
       "epoch": 0.88,
-      "eval_accuracy": 0.88,
-      "eval_f1_score": 0.8803427172582619,
-      "eval_gmean": 0.8801313530867828,
-      "eval_loss": 0.30110642313957214,
-      "eval_precision": 0.881396103896104,
-      "eval_recall": 0.88,
-      "eval_runtime": 211.6438,
-      "eval_samples_per_second": 0.472,
-      "eval_steps_per_second": 0.061,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.2766796946525574,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.4227,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 72
     },
     {
       "epoch": 0.97,
-      "grad_norm": 7.661783695220947,
+      "grad_norm": 4.84375,
       "learning_rate": 5.000000000000001e-07,
-      "loss": 0.3248,
+      "loss": 0.3856,
       "step": 80
     },
     {
       "epoch": 0.97,
-      "eval_accuracy": 0.88,
-      "eval_f1_score": 0.8803427172582619,
-      "eval_gmean": 0.8801313530867828,
-      "eval_loss": 0.29578596353530884,
-      "eval_precision": 0.881396103896104,
-      "eval_recall": 0.88,
-      "eval_runtime": 210.1348,
-      "eval_samples_per_second": 0.476,
-      "eval_steps_per_second": 0.062,
+      "eval_accuracy": 0.86,
+      "eval_f1_score": 0.8603998368013055,
+      "eval_gmean": 0.8596058209945328,
+      "eval_loss": 0.27683594822883606,
+      "eval_precision": 0.8614935064935065,
+      "eval_recall": 0.86,
+      "eval_runtime": 306.6542,
+      "eval_samples_per_second": 0.326,
+      "eval_steps_per_second": 0.042,
       "step": 80
     },
     {
       "epoch": 1.0,
       "step": 82,
       "total_flos": 1.0505428534296576e+17,
-      "train_loss": 0.5505467119013391,
-      "train_runtime": 36020.1073,
-      "train_samples_per_second": 0.146,
-      "train_steps_per_second": 0.002
+      "train_loss": 0.366200889029154,
+      "train_runtime": 19534.4865,
+      "train_samples_per_second": 0.269,
+      "train_steps_per_second": 0.004
     }
   ],
   "logging_steps": 8,