Training in progress, step 400

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "openai/whisper-base",
+  "_name_or_path": "openai/whisper-small",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,
@@ -13,17 +13,17 @@
   ],
   "bos_token_id": 50257,
   "classifier_proj_size": 256,
-  "d_model": 512,
-  "decoder_attention_heads": 8,
-  "decoder_ffn_dim": 2048,
+  "d_model": 768,
+  "decoder_attention_heads": 12,
+  "decoder_ffn_dim": 3072,
   "decoder_layerdrop": 0.0,
-  "decoder_layers": 6,
+  "decoder_layers": 12,
   "decoder_start_token_id": 50258,
   "dropout": 0.0,
-  "encoder_attention_heads": 8,
-  "encoder_ffn_dim": 2048,
+  "encoder_attention_heads": 12,
+  "encoder_ffn_dim": 3072,
   "encoder_layerdrop": 0.0,
-  "encoder_layers": 6,
+  "encoder_layers": 12,
   "eos_token_id": 50257,
   "forced_decoder_ids": null,
   "init_std": 0.02,
@@ -38,7 +38,7 @@
   "max_source_positions": 1500,
   "max_target_positions": 448,
   "model_type": "whisper",
-  "num_hidden_layers": 6,
+  "num_hidden_layers": 12,
   "num_mel_bins": 80,
   "pad_token_id": 50257,
   "scale_embedding": false,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1627c99ef6829c9afa140f61d6004d43b7286561d6ad000d987a0879db7b8323
-size 290458721
+oid sha256:f2a5990a9332646bb41af269159e24c03b63f29d49da4961a1ff533ac7a05370
+size 967102601

runs/Apr17_14-04-36_dsai-server/events.out.tfevents.1681715082.dsai-server

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:857ea98e75bea751655651985a08cc5ae88800dc2fc3a8c5e84249a8b212c711
-size 5506
+oid sha256:1457f88dc5718c01bc9f24e7519d51de34c4602085de2d19ae8523a51b0dbbd5
+size 5663

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c5d251fb523611a09de201daeeab3f95a103ab67fa22dc829dfc766b706a213e
+oid sha256:7e5781b6a2162b38531a5e6ba4e2e9bbfa39ea3b9d060f5d2bcb84134ad4c467
 size 3707