Training in progress, step 250

config.json

@@ -9,7 +9,7 @@
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.0,
+  "attention_dropout": 0.03,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
   "codevector_dim": 1024,
@@ -54,7 +54,7 @@
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.05,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
+  "hidden_dropout": 0.1,
   "hidden_size": 1280,
   "initializer_range": 0.02,
   "intermediate_size": 5120,
@@ -77,7 +77,7 @@
   "num_hidden_layers": 48,
   "num_negatives": 100,
   "output_hidden_size": 1280,
-  "pad_token_id": 143,
+  "pad_token_id": 122,
   "proj_codevector_dim": 1024,
   "tdnn_dilation": [
     1,
@@ -103,6 +103,6 @@
   "torch_dtype": "float32",
   "transformers_version": "4.31.0",
   "use_weighted_layer_sum": false,
-  "vocab_size": 146,
+  "vocab_size": 125,
   "xvector_output_dim": 512
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:538059d76281b0026c1e96eace507f48bce46d55a8724e9bcf8c0968f3e497d5
-size 3859723853
+oid sha256:e9cfae5877a8faf071e2a0360f3ec6846a3a33e0ecc8f320e74b8f5a69d14ff7
+size 3859616205

runs/Aug17_16-36-50_DESKTOP-BP9TA84/events.out.tfevents.1692282596.DESKTOP-BP9TA84

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6af30fbc7a0f98d9a289004c97988d2b255c2647ee10e1a3a69e0a3126f8c631
+size 6229

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6bcdfa4ac5dfca9e3bc6bf070856216150fd39a2f3748726f2213db38b2302a
+oid sha256:742fdb932a4e99e7cb3a7c94999a72d446d2de363e813d3785fba9671a639b12
 size 4027