Upload folder

config.json

@@ -0,0 +1,40 @@
+{
+  "_name_or_path": "/data1/speech/anhnmt2/cuongnm/EOT/Qwen2.5-0.5B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "eot_loss_weight": 1.0,
+  "hidden_act": "silu",
+  "hidden_size": 896,
+  "initializer_range": 0.02,
+  "intermediate_size": 4864,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "asr_speech_qwen2",
+  "num_attention_heads": 14,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "speech_encoder": "/data1/speech/anhnmt2/Speech2Speech/LLaMA-Omni/models/speech_encoder/whisper-medium",
+  "speech_encoder_ds_rate": 5,
+  "speech_encoder_hidden_size": 1024,
+  "speech_encoder_type": "whisper",
+  "speech_normalize": false,
+  "speech_projector_lr": null,
+  "speech_projector_type": "linear",
+  "tie_word_embeddings": true,
+  "tokenizer_model_max_length": 1024,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.0",
+  "tune_speech_projector": true,
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151665
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4f56782af56c89934eb38a1be3267a13f7aae677ec340e87184859a7f5d132f
+size 1898665066

speech_projector.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ae82fd314aa12c049d22992eb3d0adcfe39ac590803e1900b1834282c485b9a
+size 24649624

trainer_state.json

@@ -0,0 +1,70 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 1000,
+  "global_step": 42,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 6.295878887176514,
+      "learning_rate": 0.0001845190085543795,
+      "loss": 4.9016,
+      "step": 10
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 4.4954118728637695,
+      "learning_rate": 0.00012000256937760445,
+      "loss": 2.7019,
+      "step": 20
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 4.040550231933594,
+      "learning_rate": 4.3193525326884435e-05,
+      "loss": 2.1165,
+      "step": 30
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 3.2141640186309814,
+      "learning_rate": 1.2949737362087156e-06,
+      "loss": 1.5267,
+      "step": 40
+    },
+    {
+      "epoch": 3.0,
+      "step": 42,
+      "total_flos": 266392691539968.0,
+      "train_loss": 2.760472910744803,
+      "train_runtime": 117.3664,
+      "train_samples_per_second": 1.431,
+      "train_steps_per_second": 0.358
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 42,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 266392691539968.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}