hf-internal-testing
/

tiny-random-m2m_100

stas commited on Apr 12, 2022

Commit

3b3cd4a

1 Parent(s): b24ccc7

update max_position_embeddings and layerdrop probabilities

Files changed (1) hide show

config.json CHANGED Viewed

@@ -6,19 +6,19 @@
   "d_model": 16,
   "decoder_attention_heads": 4,
   "decoder_ffn_dim": 4,
-  "decoder_layerdrop": 0.0,
   "decoder_layers": 2,
   "decoder_start_token_id": 2,
   "dropout": 0.1,
   "encoder_attention_heads": 4,
   "encoder_ffn_dim": 4,
-  "encoder_layerdrop": 0.0,
   "encoder_layers": 2,
   "eos_token_id": 2,
   "gradient_checkpointing": false,
   "init_std": 0.02,
   "is_encoder_decoder": true,
-  "max_position_embeddings": 20,
   "model_type": "m2m_100",
   "num_hidden_layers": 2,
   "pad_token_id": 1,

   "d_model": 16,
   "decoder_attention_heads": 4,
   "decoder_ffn_dim": 4,
+  "decoder_layerdrop": 0.05,
   "decoder_layers": 2,
   "decoder_start_token_id": 2,
   "dropout": 0.1,
   "encoder_attention_heads": 4,
   "encoder_ffn_dim": 4,
+  "encoder_layerdrop": 0.05,
   "encoder_layers": 2,
   "eos_token_id": 2,
   "gradient_checkpointing": false,
   "init_std": 0.02,
   "is_encoder_decoder": true,
+  "max_position_embeddings": 512,
   "model_type": "m2m_100",
   "num_hidden_layers": 2,
   "pad_token_id": 1,