Training in progress, step 500

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "_name_or_path": "facebook/wav2vec2-base",
   "activation_dropout": 0.0,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,
@@ -12,9 +12,9 @@
   "attention_dropout": 0.1,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "codevector_dim": 768,
+  "codevector_dim": 256,
   "contrastive_logits_temperature": 0.1,
-  "conv_bias": true,
+  "conv_bias": false,
   "conv_dim": [
     512,
     512,
@@ -45,41 +45,50 @@
   "ctc_loss_reduction": "mean",
   "ctc_zero_infinity": false,
   "diversity_loss_weight": 0.1,
-  "do_stable_layer_norm": true,
+  "do_stable_layer_norm": false,
   "eos_token_id": 2,
   "feat_extract_activation": "gelu",
-  "feat_extract_dropout": 0.0,
-  "feat_extract_norm": "layer",
+  "feat_extract_norm": "group",
   "feat_proj_dropout": 0.1,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
-  "gradient_checkpointing": false,
+  "freeze_feat_extract_train": true,
   "hidden_act": "gelu",
   "hidden_dropout": 0.1,
-  "hidden_size": 1024,
+  "hidden_size": 768,
   "initializer_range": 0.02,
-  "intermediate_size": 4096,
+  "intermediate_size": 3072,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.1,
+  "layerdrop": 0.0,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
   "mask_feature_length": 10,
   "mask_feature_min_masks": 0,
   "mask_feature_prob": 0.0,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
-  "mask_time_prob": 0.075,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
   "model_type": "wav2vec2",
+  "no_mask_channel_overlap": false,
+  "no_mask_time_overlap": false,
   "num_adapter_layers": 3,
-  "num_attention_heads": 16,
+  "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": 24,
+  "num_hidden_layers": 12,
   "num_negatives": 100,
-  "output_hidden_size": 1024,
+  "output_hidden_size": 768,
   "pad_token_id": 42,
-  "proj_codevector_dim": 768,
+  "proj_codevector_dim": 256,
   "tdnn_dilation": [
     1,
     2,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d39798d8dde8e1b3c3e37b8bff1befebb2070ac082074e67722733a9bf161dcc
-size 1261991980
+oid sha256:7a89140bf7a483b3d878d0229f9b9d285285e3c0754e93642f7d2bcc8e362561
+size 377651116

runs/Apr20_16-01-47_1c16ff80eb9d/events.out.tfevents.1713629031.1c16ff80eb9d.191.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f07039cee7e9d05947789de5566cc76bbad72ba08ab1df978e09c96546e63e7
+size 7864

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b69c73def89e0d3a1a6b714dd955272aa39531e6d18a7525366ee5479c6b380
+oid sha256:fc50023c4a0087e861633ae8c447bdf47dffab29e4fe5134facbeaa88cf4a521
 size 5048