Training in progress, epoch 1

config.json

@@ -9,18 +9,18 @@
   "begin_suppress_tokens": null,
   "bos_token_id": 50257,
   "classifier_proj_size": 256,
-  "d_model": 384,
-  "decoder_attention_heads": 6,
-  "decoder_ffn_dim": 1536,
+  "d_model": 768,
+  "decoder_attention_heads": 12,
+  "decoder_ffn_dim": 3072,
   "decoder_layerdrop": 0.0,
-  "decoder_layers": 4,
+  "decoder_layers": 12,
   "decoder_start_token_id": 50258,
   "dropout": 0.0,
   "dtype": "float32",
-  "encoder_attention_heads": 6,
-  "encoder_ffn_dim": 1536,
+  "encoder_attention_heads": 12,
+  "encoder_ffn_dim": 3072,
   "encoder_layerdrop": 0.0,
-  "encoder_layers": 4,
+  "encoder_layers": 12,
   "eos_token_id": 50257,
   "forced_decoder_ids": [
     [
@@ -49,7 +49,7 @@
   "max_target_positions": 448,
   "median_filter_width": 7,
   "model_type": "whisper",
-  "num_hidden_layers": 4,
+  "num_hidden_layers": 12,
   "num_mel_bins": 80,
   "pad_token_id": 50257,
   "scale_embedding": false,

generation_config.json

@@ -1,27 +1,43 @@
 {
   "alignment_heads": [
     [
-      2,
-      2
+      5,
+      3
+    ],
+    [
+      5,
+      9
     ],
     [
-      3,
+      8,
       0
     ],
     [
-      3,
-      2
+      8,
+      4
     ],
     [
-      3,
-      3
+      8,
+      7
     ],
     [
-      3,
-      4
+      8,
+      8
+    ],
+    [
+      9,
+      0
+    ],
+    [
+      9,
+      7
+    ],
+    [
+      9,
+      9
     ],
     [
-      3,
+      10,
       5
     ]
   ],

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d563460e05d9374542048a75ecf93c0ce522e85e8d270432bd709a56779275d1
-size 151061672
+oid sha256:8e7580e62bb03164905ffa6be95a9863a9a5abf54533e773b214f846a58a411c
+size 966995080

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff39a4739f850346834362756ffe52cda757cf6512ee98bf419f5373eda8d90c
+oid sha256:3e6d12cc6233cb8c46314e1437089e0fcbdfc8242f4a0d436fc52d3c37d80a9e
 size 6033