Training in progress, step 500

adapter_config.json

@@ -0,0 +1,41 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3-0.6B",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.0",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3b33f551146fc1725dd90ee0da41d2e15f8ef15cd21800ef4e3d512564d9ff5
+size 36715216

config.json

@@ -156,24 +156,24 @@
     ]
   ],
   "encoder_dim": 1280,
-  "freeze_projector": false,
+  "freeze_projector": true,
   "inference_warmup_tokens": 10,
   "label_smoothing": 0.0,
   "length_penalty": 1.0,
   "llm_dim": 1024,
-  "lora_alpha": 32,
+  "lora_alpha": 128,
   "lora_dropout": 0.0,
-  "lora_rank": 8,
+  "lora_rank": 64,
   "lora_target_modules": [
     "q_proj",
     "v_proj"
   ],
-  "mask_feature_length": 27,
-  "mask_feature_min_masks": 2,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
   "mask_feature_prob": 0.0,
-  "mask_time_length": 100,
+  "mask_time_length": 10,
   "mask_time_min_masks": 2,
-  "mask_time_prob": 0.0,
+  "mask_time_prob": 0.05,
   "max_new_tokens": 256,
   "min_new_tokens": 0,
   "model_dtype": "bfloat16",
@@ -183,6 +183,7 @@
   "num_experts": 4,
   "num_experts_per_tok": 2,
   "pipeline_tag": "automatic-speech-recognition",
+  "pretrained_model_path": "mazesmazes/tiny-audio",
   "projector_dropout": 0.0,
   "projector_hidden_dim": null,
   "projector_init_std": 0.02,
@@ -262,7 +263,7 @@
   "text_model_id": "Qwen/Qwen3-0.6B",
   "transformers_version": "5.0.0.dev0",
   "use_cache": false,
-  "use_lora": false,
+  "use_lora": true,
   "use_specaugment": false,
   "user_prompt": "Please transcribe this English audio into text: <audio>",
   "vocab_size": 151670

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66e4fd5269ddd0a85cab49962bc693e29b8e66f64f6a5ac11347bafc2311b7e3
+oid sha256:50389fcf6939a2d04fa7fb7e239f98f27e46ad21c06f320b15fb023f3550c88d
 size 5201