End of training

README.md

@@ -1,5 +1,6 @@
 ---
-base_model: VenkatManda/BertMiniV3
+license: mit
+base_model: VenkatManda/VV2
 tags:
 - generated_from_trainer
 datasets:
@@ -14,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # BertMiniV3
 
-This model is a fine-tuned version of [VenkatManda/BertMiniV3](https://huggingface.co/VenkatManda/BertMiniV3) on the squad_v2 dataset.
+This model is a fine-tuned version of [VenkatManda/VV2](https://huggingface.co/VenkatManda/VV2) on the squad_v2 dataset.
 
 ## Model description
 

all_results.json

@@ -1,19 +1,19 @@
 {
-    "eval_HasAns_exact": 43.68252698920432,
-    "eval_HasAns_f1": 51.86559486275948,
+    "eval_HasAns_exact": 22.97081167532987,
+    "eval_HasAns_f1": 29.82730501295977,
     "eval_HasAns_total": 5002,
-    "eval_NoAns_exact": 50.40016006402561,
-    "eval_NoAns_f1": 50.40016006402561,
+    "eval_NoAns_exact": 61.644657863145255,
+    "eval_NoAns_f1": 61.644657863145255,
     "eval_NoAns_total": 4998,
-    "eval_best_exact": 50.02,
+    "eval_best_exact": 50.0,
     "eval_best_exact_thresh": 0.0,
-    "eval_best_f1": 51.163646740828945,
+    "eval_best_f1": 50.00666666666667,
     "eval_best_f1_thresh": 0.0,
-    "eval_exact": 47.04,
-    "eval_f1": 51.13317055035233,
-    "eval_runtime": 840.5387,
+    "eval_exact": 42.3,
+    "eval_f1": 45.72961796748245,
+    "eval_runtime": 188.1784,
     "eval_samples": 10000,
-    "eval_samples_per_second": 11.897,
-    "eval_steps_per_second": 1.487,
+    "eval_samples_per_second": 53.141,
+    "eval_steps_per_second": 6.643,
     "eval_total": 10000
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "VenkatManda/BertMiniV3",
+  "_name_or_path": "VenkatManda/VV2",
   "architectures": [
     "BertForQuestionAnswering"
   ],
@@ -7,14 +7,14 @@
   "classifier_dropout": null,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 256,
+  "hidden_size": 128,
   "initializer_range": 0.02,
-  "intermediate_size": 1024,
+  "intermediate_size": 512,
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
   "model_type": "bert",
-  "num_attention_heads": 4,
-  "num_hidden_layers": 4,
+  "num_attention_heads": 2,
+  "num_hidden_layers": 2,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fbb2d9ac216fdceb2d22af1a1bf915dcbb8bed7ad0cd5a0a022cd02810dd2ed9
-size 46353720
+oid sha256:3ef0f25abc2491c1b8c728a50e3ca6760579f5e0052f19a76618aad8a8a3273d
+size 46232490

eval_results.json

@@ -1,19 +1,19 @@
 {
-    "eval_HasAns_exact": 43.68252698920432,
-    "eval_HasAns_f1": 51.86559486275948,
+    "eval_HasAns_exact": 22.97081167532987,
+    "eval_HasAns_f1": 29.82730501295977,
     "eval_HasAns_total": 5002,
-    "eval_NoAns_exact": 50.40016006402561,
-    "eval_NoAns_f1": 50.40016006402561,
+    "eval_NoAns_exact": 61.644657863145255,
+    "eval_NoAns_f1": 61.644657863145255,
     "eval_NoAns_total": 4998,
-    "eval_best_exact": 50.02,
+    "eval_best_exact": 50.0,
     "eval_best_exact_thresh": 0.0,
-    "eval_best_f1": 51.163646740828945,
+    "eval_best_f1": 50.00666666666667,
     "eval_best_f1_thresh": 0.0,
-    "eval_exact": 47.04,
-    "eval_f1": 51.13317055035233,
-    "eval_runtime": 840.5387,
+    "eval_exact": 42.3,
+    "eval_f1": 45.72961796748245,
+    "eval_runtime": 188.1784,
     "eval_samples": 10000,
-    "eval_samples_per_second": 11.897,
-    "eval_steps_per_second": 1.487,
+    "eval_samples_per_second": 53.141,
+    "eval_steps_per_second": 6.643,
     "eval_total": 10000
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82c72e04ef8bbadbe2a44bc64bd95af4b21889cc322994d3200741ad21a39da9
-size 44429248
+oid sha256:a91163295a0d55760e2792d71ca80fcdb4bb57e7ffbe659ad725b5bcfe880c62
+size 17483080