Training in progress - step 5000

asr_modeling.py

@@ -573,17 +573,21 @@ class ASRModel(PreTrainedModel, GenerationMixin):
         )
 
         # Generate using language model
+        # Pass both input_ids and inputs_embeds so repetition_penalty works correctly
+        # (it needs input_ids to track which tokens have been used)
         output = self.language_model.generate(
+            input_ids=input_ids,
             inputs_embeds=inputs_embeds,
             attention_mask=attention_mask,
             generation_config=self.generation_config,
             **generate_kwargs,
         )
 
-        # When using inputs_embeds without input_ids, generate returns only new tokens
-        if isinstance(output, torch.Tensor):
-            return output
-        return output.sequences
+        # When using inputs_embeds with input_ids, generate returns full sequence
+        # Strip the input tokens to return only generated tokens
+        sequences = output if isinstance(output, torch.Tensor) else output.sequences
+        input_len = input_ids.shape[1]
+        return sequences[:, input_len:]
 
     def generate_streaming(
         self,

config.json

@@ -274,7 +274,7 @@
   "qformer_num_heads": 16,
   "qformer_num_layers": 2,
   "qformer_window_size": 15,
-  "repetition_penalty": 1.0,
+  "repetition_penalty": 1.05,
   "router_aux_loss_coef": 0.01,
   "system_prompt": "",
   "temperature": 0.7,

generation_config.json

@@ -11,7 +11,7 @@
   "no_repeat_ngram_size": 0,
   "num_beams": 1,
   "pad_token_id": 151643,
-  "repetition_penalty": 1.0,
+  "repetition_penalty": 1.05,
   "temperature": 0.7,
   "transformers_version": "5.0.0.dev0",
   "use_cache": true

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8d27e4a7c907ced9bf93828c060f517ab52e14aa1d2d507a1ce23f8ae3f9435
+oid sha256:e1c29578f6e4473b5f6a25ba03515832cfc1c5698f516d02f7758722d09b7065
 size 58732960