Training in progress, step 200

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/mms-1b",
+  "_name_or_path": "facebook/mms-300m",
   "activation_dropout": 0.0,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,
@@ -12,7 +12,7 @@
   "attention_dropout": 0.1,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "codevector_dim": 1024,
+  "codevector_dim": 768,
   "contrastive_logits_temperature": 0.1,
   "conv_bias": true,
   "conv_dim": [
@@ -56,9 +56,9 @@
   "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout": 0.1,
-  "hidden_size": 1280,
+  "hidden_size": 1024,
   "initializer_range": 0.02,
-  "intermediate_size": 5120,
+  "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
   "layerdrop": 0.0,
   "mask_feature_length": 10,
@@ -75,11 +75,11 @@
   "num_conv_pos_embedding_groups": 16,
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
-  "num_hidden_layers": 48,
+  "num_hidden_layers": 24,
   "num_negatives": 100,
-  "output_hidden_size": 1280,
+  "output_hidden_size": 1024,
   "pad_token_id": 33,
-  "proj_codevector_dim": 1024,
+  "proj_codevector_dim": 768,
   "tdnn_dilation": [
     1,
     2,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94663caf9113a0f82c6ee79d873972394d7830508c5c7c7de090daf512b98b1e
-size 3850275464
+oid sha256:300204f0afd9443aa6fcb8c15162630bb17e8d5ce648328037bd2129bdd7e4d0
+size 1261955080

runs/Jun12_21-37-33_103f35af7d2e/events.out.tfevents.1718228358.103f35af7d2e.2293.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b7dc091665a254cacdbc437ad8169a166b3121d7fa64948543cd25bcecbac2f
+size 14004

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aa23c7ce281022431199386fff78b4eab4ab67a2dfe22b4e065e631d7974fd2f
+oid sha256:479edb3bdc92f031a3b9810004e4a6898e211066e75ffcdfcff328760ac2dfac
 size 5048