Training in progress, step 600

config.json

@@ -8,7 +8,7 @@
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.05,
+  "attention_dropout": 0.1,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
   "codevector_dim": 768,
@@ -52,8 +52,9 @@
   "feat_proj_dropout": 0.1,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
+  "gradient_checkpointing": false,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.05,
+  "hidden_dropout": 0.1,
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 4096,
@@ -81,7 +82,7 @@
   "num_conv_pos_embedding_groups": 16,
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
-  "num_hidden_layers": 24,
+  "num_hidden_layers": 12,
   "num_negatives": 100,
   "output_hidden_size": 1024,
   "pad_token_id": 69,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c4347cc7e50cb0569d9e495fc040d0799e74d98539486daf11ab943ce141c54
-size 1262197037
+oid sha256:c2381cbd87df45bd0895d16a04099dac61ab0643b2e2e174a01b2180b4f639dc
+size 657511405

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:606a218a459c75cce09de7275cf74d1bb9f36e7119346742ed22a880db485eb7
+oid sha256:6a7b7ccdca35fdb287f013757d84481fabc2462920fd8131999514145c7bb715
 size 3579