Training in progress, epoch 1

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/nllb-200-1.3B",
+  "_name_or_path": "facebook/nllb-200-distilled-600M",
   "activation_dropout": 0.0,
   "activation_function": "relu",
   "architectures": [
@@ -9,25 +9,26 @@
   "bos_token_id": 0,
   "d_model": 1024,
   "decoder_attention_heads": 16,
-  "decoder_ffn_dim": 8192,
+  "decoder_ffn_dim": 4096,
   "decoder_layerdrop": 0,
-  "decoder_layers": 24,
+  "decoder_layers": 12,
   "decoder_start_token_id": 2,
   "dropout": 0.1,
   "encoder_attention_heads": 16,
-  "encoder_ffn_dim": 8192,
+  "encoder_ffn_dim": 4096,
   "encoder_layerdrop": 0,
-  "encoder_layers": 24,
+  "encoder_layers": 12,
   "eos_token_id": 2,
   "init_std": 0.02,
   "is_encoder_decoder": true,
   "max_length": 200,
   "max_position_embeddings": 1024,
   "model_type": "m2m_100",
-  "num_hidden_layers": 24,
+  "num_hidden_layers": 12,
   "pad_token_id": 1,
   "scale_embedding": true,
-  "torch_dtype": "float16",
+  "tokenizer_class": "NllbTokenizer",
+  "torch_dtype": "float32",
   "transformers_version": "4.39.3",
   "use_cache": true,
   "vocab_size": 256206

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0576dc3f26cdd00433ba4d96068d992ae40ffb0868e0f4b0b854cda0e44ab1c4
-size 2741396584
+oid sha256:d05416ac70afadc3511eb82b32cdbb9d44e1102f0ce01b68d2076b6ff0394a66
+size 2460354912

runs/Jun01_23-18-08_628c2640942c/events.out.tfevents.1717283890.628c2640942c.23.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01769003e1fc3be0069e0476361e7323cdaf313a525e6c87e24b265d8e8e797c
+size 10889

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79d46f04d5542e99359568acc8c450715c649e99e6f3c64bbe6eb5c49e0dda79
+oid sha256:a16f4d12ed1a2e80c89c4f6d3a609836bd7fc47ed22dfe36debf42b960503106
 size 5112