Training in progress, step 600

config.json

@@ -8,7 +8,7 @@
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.1,
+  "attention_dropout": 0,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
   "codevector_dim": 768,
@@ -54,12 +54,12 @@
   "final_dropout": 0.0,
   "gradient_checkpointing": false,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.1,
+  "hidden_dropout": 0,
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.2,
+  "layerdrop": 0.15,
   "mask_channel_length": 10,
   "mask_channel_min_space": 1,
   "mask_channel_other": 0.0,
@@ -82,7 +82,7 @@
   "num_conv_pos_embedding_groups": 16,
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
-  "num_hidden_layers": 12,
+  "num_hidden_layers": 24,
   "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 sha256:6a667ab7c92594bc17fc0aa90bafb12dc7df4bc42a09423557e9a07c4b4ebae5
-size 657511405
+oid sha256:c8644d92469c5674bdf446db7c4469381a63b850a4d7810a2602f4837fd44d71
+size 1262197037

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a7b7ccdca35fdb287f013757d84481fabc2462920fd8131999514145c7bb715
+oid sha256:1348f423dcb2881b15214d1014360addd934853b4093fdac4cf79c753eecc07f
 size 3579