Training in progress, epoch 0

config.json

@@ -55,9 +55,9 @@
   "final_dropout": 0.1,
   "hidden_act": "gelu",
   "hidden_dropout": 0.1,
-  "hidden_size": 128,
+  "hidden_size": 384,
   "initializer_range": 0.02,
-  "intermediate_size": 512,
+  "intermediate_size": 1536,
   "layer_norm_eps": 1e-05,
   "layerdrop": 0.1,
   "mask_feature_length": 10,
@@ -68,13 +68,13 @@
   "mask_time_prob": 0.05,
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,
-  "num_attention_heads": 4,
+  "num_attention_heads": 12,
   "num_codevector_groups": 2,
   "num_codevectors_per_group": 320,
   "num_conv_pos_embedding_groups": 16,
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
-  "num_hidden_layers": 2,
+  "num_hidden_layers": 6,
   "num_negatives": 100,
   "output_hidden_size": 768,
   "pad_token_id": 49,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4cebc332b9ed16a1dd472950e8b307a00578872de8b946d43f24cd62f165752f
-size 19214560
+oid sha256:32c5df6b29386e3581bae9e108f81b3403c666e7490422494e155844b236d9ef
+size 65000248

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1bfafd9ee79459acfdc7195ceefdf3f625ca40437e586859a4bee6076728d433
+oid sha256:30ee42f67045bba106f47f0f6f5f062c3e5114c16ce8a08158710f7262318745
 size 5176