Training in progress, step 100

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "openai/whisper-base.en",
+  "_name_or_path": "openai/whisper-small.en",
   "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": 50257,
   "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": 50256,
   "forced_decoder_ids": null,
   "init_std": 0.02,
@@ -39,13 +39,13 @@
   "max_target_positions": 448,
   "median_filter_width": 7,
   "model_type": "whisper",
-  "num_hidden_layers": 6,
+  "num_hidden_layers": 12,
   "num_mel_bins": 80,
   "pad_token_id": 50256,
   "scale_embedding": false,
   "suppress_tokens": [],
   "torch_dtype": "float32",
-  "transformers_version": "4.38.0.dev0",
+  "transformers_version": "4.41.0",
   "use_cache": true,
   "use_weighted_layer_sum": false,
   "vocab_size": 51864

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05b4594d793f2f1e9ce21e71db7d80799da63426c9e0f93b35b90886224a0486
-size 290401888
+oid sha256:bc39f2073821d5e5adcb02e26cad8cf85ecc008b1a993e3aa59a4fc22071f9a5
+size 966992008

runs/May22_05-20-14_YanYue-MS-7C06/events.out.tfevents.1716355728.YanYue-MS-7C06

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1e0dcdb3b59598c389554a6f7de85bde0ec4157b3978ea9960ae3b619ee3d0f
+size 5600

runs/May22_14-24-02_z390-a/events.out.tfevents.1716388047.z390-a.1171.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52fb1bac752f92b5121c10f138df8361e6c074f1896dd5530f6e0767f6dda21a
+size 6732

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b4a6d12b1fe69f8269a040f3fc35168c5fb24c41b893dd22bf8395831555908
-size 4856
+oid sha256:b6984e4ad7bd525814bf4577fdf8a8f54dcfa5d5eb5d434e421526ed2fe5043e
+size 5240