End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: gokulsrinivasagan/distilbert_lda_5_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: distilbert_lda_5_v1_qnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QNLI
+      type: glue
+      args: qnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8434925864909391
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # distilbert_lda_5_v1_qnli
 
-This model is a fine-tuned version of [gokulsrinivasagan/distilbert_lda_5_v1](https://huggingface.co/gokulsrinivasagan/distilbert_lda_5_v1) on an unknown dataset.
+This model is a fine-tuned version of [gokulsrinivasagan/distilbert_lda_5_v1](https://huggingface.co/gokulsrinivasagan/distilbert_lda_5_v1) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6610
-- Accuracy: 0.8122
+- Loss: 0.3706
+- Accuracy: 0.8435
 
 ## Model description
 

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 12.0,
-    "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.691364049911499,
-    "eval_runtime": 2.3751,
+    "epoch": 7.0,
+    "eval_accuracy": 0.8434925864909391,
+    "eval_loss": 0.3705822825431824,
+    "eval_runtime": 2.2775,
     "eval_samples": 5463,
-    "eval_samples_per_second": 2300.09,
-    "eval_steps_per_second": 9.263,
-    "total_flos": 8.325019642455245e+16,
-    "train_loss": 0.6955604677277852,
-    "train_runtime": 1216.889,
+    "eval_samples_per_second": 2398.7,
+    "eval_steps_per_second": 9.66,
+    "total_flos": 4.856261458098893e+16,
+    "train_loss": 0.23536453379986594,
+    "train_runtime": 707.5282,
     "train_samples": 104743,
-    "train_samples_per_second": 4303.72,
-    "train_steps_per_second": 16.846
+    "train_samples_per_second": 7402.037,
+    "train_steps_per_second": 28.974
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 12.0,
-    "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.691364049911499,
-    "eval_runtime": 2.3751,
+    "epoch": 7.0,
+    "eval_accuracy": 0.8434925864909391,
+    "eval_loss": 0.3705822825431824,
+    "eval_runtime": 2.2775,
     "eval_samples": 5463,
-    "eval_samples_per_second": 2300.09,
-    "eval_steps_per_second": 9.263
+    "eval_samples_per_second": 2398.7,
+    "eval_steps_per_second": 9.66
 }

logs/events.out.tfevents.1733310297.ki-g0008.1206098.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a850f9b8d3524981aa3633e0fcfa3eb18401343605ccf4b0563eb2f36df97663
+size 411

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 12.0,
-    "total_flos": 8.325019642455245e+16,
-    "train_loss": 0.6955604677277852,
-    "train_runtime": 1216.889,
+    "epoch": 7.0,
+    "total_flos": 4.856261458098893e+16,
+    "train_loss": 0.23536453379986594,
+    "train_runtime": 707.5282,
     "train_samples": 104743,
-    "train_samples_per_second": 4303.72,
-    "train_steps_per_second": 16.846
+    "train_samples_per_second": 7402.037,
+    "train_steps_per_second": 28.974
 }

trainer_state.json

@@ -1,213 +1,133 @@
 {
-  "best_metric": 0.691364049911499,
-  "best_model_checkpoint": "distilbert_lda_5_v1_qnli/checkpoint-2870",
-  "epoch": 12.0,
+  "best_metric": 0.3705822825431824,
+  "best_model_checkpoint": "distilbert_lda_5_v1_qnli/checkpoint-820",
+  "epoch": 7.0,
   "eval_steps": 500,
-  "global_step": 4920,
+  "global_step": 2870,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 0.16467422246932983,
-      "learning_rate": 0.00098,
-      "loss": 0.7218,
+      "grad_norm": 4.038874626159668,
+      "learning_rate": 4.9e-05,
+      "loss": 0.497,
       "step": 410
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6934226751327515,
-      "eval_runtime": 2.347,
-      "eval_samples_per_second": 2327.7,
-      "eval_steps_per_second": 9.374,
+      "eval_accuracy": 0.8303130148270181,
+      "eval_loss": 0.38713347911834717,
+      "eval_runtime": 2.272,
+      "eval_samples_per_second": 2404.487,
+      "eval_steps_per_second": 9.683,
       "step": 410
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.2183358371257782,
-      "learning_rate": 0.00096,
-      "loss": 0.6932,
+      "grad_norm": 4.284979820251465,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3635,
       "step": 820
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6935070157051086,
-      "eval_runtime": 2.3054,
-      "eval_samples_per_second": 2369.647,
-      "eval_steps_per_second": 9.543,
+      "eval_accuracy": 0.8434925864909391,
+      "eval_loss": 0.3705822825431824,
+      "eval_runtime": 2.3214,
+      "eval_samples_per_second": 2353.364,
+      "eval_steps_per_second": 9.477,
       "step": 820
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.1270582526922226,
-      "learning_rate": 0.00094,
-      "loss": 0.6932,
+      "grad_norm": 4.197302341461182,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2745,
       "step": 1230
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 2.3726,
-      "eval_samples_per_second": 2302.498,
-      "eval_steps_per_second": 9.272,
+      "eval_accuracy": 0.8376349990847519,
+      "eval_loss": 0.37501493096351624,
+      "eval_runtime": 2.303,
+      "eval_samples_per_second": 2372.144,
+      "eval_steps_per_second": 9.553,
       "step": 1230
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.01812651753425598,
-      "learning_rate": 0.00092,
-      "loss": 0.6931,
+      "grad_norm": 6.512701511383057,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.1968,
       "step": 1640
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 2.3609,
-      "eval_samples_per_second": 2313.915,
-      "eval_steps_per_second": 9.318,
+      "eval_accuracy": 0.8308621636463481,
+      "eval_loss": 0.45666348934173584,
+      "eval_runtime": 2.3076,
+      "eval_samples_per_second": 2367.403,
+      "eval_steps_per_second": 9.534,
       "step": 1640
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.05285927280783653,
-      "learning_rate": 0.0009000000000000001,
-      "loss": 0.6932,
+      "grad_norm": 4.087521553039551,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1394,
       "step": 2050
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6934226751327515,
-      "eval_runtime": 2.4032,
-      "eval_samples_per_second": 2273.215,
-      "eval_steps_per_second": 9.154,
+      "eval_accuracy": 0.8321435108914516,
+      "eval_loss": 0.5239177346229553,
+      "eval_runtime": 2.333,
+      "eval_samples_per_second": 2341.607,
+      "eval_steps_per_second": 9.43,
       "step": 2050
     },
     {
       "epoch": 6.0,
-      "grad_norm": 0.05455218255519867,
-      "learning_rate": 0.00088,
-      "loss": 0.6932,
+      "grad_norm": 8.565244674682617,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1001,
       "step": 2460
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 2.399,
-      "eval_samples_per_second": 2277.166,
-      "eval_steps_per_second": 9.17,
+      "eval_accuracy": 0.8341570565623284,
+      "eval_loss": 0.5522398352622986,
+      "eval_runtime": 2.2746,
+      "eval_samples_per_second": 2401.711,
+      "eval_steps_per_second": 9.672,
       "step": 2460
     },
     {
       "epoch": 7.0,
-      "grad_norm": 0.019881071522831917,
-      "learning_rate": 0.00086,
-      "loss": 0.6931,
+      "grad_norm": 2.2605297565460205,
+      "learning_rate": 4.3e-05,
+      "loss": 0.0764,
       "step": 2870
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.691364049911499,
-      "eval_runtime": 2.3314,
-      "eval_samples_per_second": 2343.275,
-      "eval_steps_per_second": 9.437,
+      "eval_accuracy": 0.8121911037891268,
+      "eval_loss": 0.6609876155853271,
+      "eval_runtime": 2.3532,
+      "eval_samples_per_second": 2321.488,
+      "eval_steps_per_second": 9.349,
       "step": 2870
     },
     {
-      "epoch": 8.0,
-      "grad_norm": 0.13051092624664307,
-      "learning_rate": 0.00084,
-      "loss": 0.693,
-      "step": 3280
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6934226751327515,
-      "eval_runtime": 2.3863,
-      "eval_samples_per_second": 2289.292,
-      "eval_steps_per_second": 9.219,
-      "step": 3280
-    },
-    {
-      "epoch": 9.0,
-      "grad_norm": 0.05455218255519867,
-      "learning_rate": 0.00082,
-      "loss": 0.6932,
-      "step": 3690
-    },
-    {
-      "epoch": 9.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 2.3723,
-      "eval_samples_per_second": 2302.849,
-      "eval_steps_per_second": 9.274,
-      "step": 3690
-    },
-    {
-      "epoch": 10.0,
-      "grad_norm": 0.057659585028886795,
-      "learning_rate": 0.0008,
-      "loss": 0.6932,
-      "step": 4100
-    },
-    {
-      "epoch": 10.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6934226751327515,
-      "eval_runtime": 2.3678,
-      "eval_samples_per_second": 2307.24,
-      "eval_steps_per_second": 9.291,
-      "step": 4100
-    },
-    {
-      "epoch": 11.0,
-      "grad_norm": 0.09045995026826859,
-      "learning_rate": 0.0007800000000000001,
-      "loss": 0.6932,
-      "step": 4510
-    },
-    {
-      "epoch": 11.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 2.3728,
-      "eval_samples_per_second": 2302.381,
-      "eval_steps_per_second": 9.272,
-      "step": 4510
-    },
-    {
-      "epoch": 12.0,
-      "grad_norm": 0.01812651753425598,
-      "learning_rate": 0.00076,
-      "loss": 0.6932,
-      "step": 4920
-    },
-    {
-      "epoch": 12.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6933804750442505,
-      "eval_runtime": 2.3791,
-      "eval_samples_per_second": 2296.263,
-      "eval_steps_per_second": 9.247,
-      "step": 4920
-    },
-    {
-      "epoch": 12.0,
-      "step": 4920,
-      "total_flos": 8.325019642455245e+16,
-      "train_loss": 0.6955604677277852,
-      "train_runtime": 1216.889,
-      "train_samples_per_second": 4303.72,
-      "train_steps_per_second": 16.846
+      "epoch": 7.0,
+      "step": 2870,
+      "total_flos": 4.856261458098893e+16,
+      "train_loss": 0.23536453379986594,
+      "train_runtime": 707.5282,
+      "train_samples_per_second": 7402.037,
+      "train_steps_per_second": 28.974
     }
   ],
   "logging_steps": 1,
@@ -236,7 +156,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8.325019642455245e+16,
+  "total_flos": 4.856261458098893e+16,
   "train_batch_size": 256,
   "trial_name": null,
   "trial_params": null