Update config.json

config.json

@@ -1,39 +1,34 @@
 {
-  "_attn_implementation_autoset": true,
-  "_name_or_path": "eustlb/moonshine",
-  "apply_spec_augment": false,
   "architectures": [
-    "MoonshineModel"
+    "MoonshineForConditionalGeneration"
   ],
   "attention_bias": false,
   "attention_dropout": 0.0,
   "bos_token_id": 1,
   "decoder_hidden_act": "silu",
+  "decoder_num_attention_heads": 8,
+  "decoder_num_hidden_layers": 6,
+  "decoder_num_key_value_heads": 8,
   "decoder_start_token_id": 1,
   "encoder_hidden_act": "gelu",
+  "encoder_num_attention_heads": 8,
+  "encoder_num_hidden_layers": 6,
+  "encoder_num_key_value_heads": 8,
   "eos_token_id": 2,
-  "ff_mult": 4,
   "hidden_size": 288,
   "initializer_range": 0.02,
-  "intermediate_size": null,
+  "intermediate_size": 1152,
   "is_encoder_decoder": true,
-  "layer_norm_eps": 1e-05,
-  "mask_feature_length": 10,
-  "mask_feature_min_masks": 0,
-  "mask_feature_prob": 0.0,
-  "mask_time_length": 10,
-  "mask_time_min_masks": 2,
-  "mask_time_prob": 0.05,
-  "max_position_embeddings": 2048,
+  "max_position_embeddings": 512,
   "model_type": "moonshine",
   "num_attention_heads": 8,
   "num_hidden_layers": 6,
   "num_key_value_heads": 8,
-  "partial_rotary_factor": 0.5,
-  "qk_layernorm": false,
+  "partial_rotary_factor": 0.9,
   "rope_scaling": null,
   "rope_theta": 10000.0,
-  "transformers_version": "4.47.0.dev0",
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.0.dev0",
   "use_cache": true,
   "vocab_size": 32768
-}
+}