Model save

README.md

@@ -0,0 +1,64 @@
+---
+base_model: google/gemma-2-2b-it
+datasets:
+- generator
+library_name: peft
+license: gemma
+tags:
+- trl
+- sft
+- generated_from_trainer
+model-index:
+- name: gemma-2-2b-it-lora-jap-en
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# gemma-2-2b-it-lora-jap-en
+
+This model is a fine-tuned version of [google/gemma-2-2b-it](https://huggingface.co/google/gemma-2-2b-it) on the generator dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- total_eval_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 3
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.44.2
+- Pytorch 2.3.0+cu121
+- Datasets 3.0.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.909090909090909,
+    "total_flos": 88796276195328.0,
+    "train_loss": 2.5086214343706765,
+    "train_runtime": 154.8962,
+    "train_samples": 14593,
+    "train_samples_per_second": 4.997,
+    "train_steps_per_second": 0.31
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.909090909090909,
+    "total_flos": 88796276195328.0,
+    "train_loss": 2.5086214343706765,
+    "train_runtime": 154.8962,
+    "train_samples": 14593,
+    "train_samples_per_second": 4.997,
+    "train_steps_per_second": 0.31
+}

trainer_state.json

@@ -0,0 +1,70 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.909090909090909,
+  "eval_steps": 500,
+  "global_step": 48,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.6060606060606061,
+      "grad_norm": 2.365993915131312,
+      "learning_rate": 0.0001934016108732548,
+      "loss": 4.9025,
+      "step": 10
+    },
+    {
+      "epoch": 1.2121212121212122,
+      "grad_norm": 0.5010752859822769,
+      "learning_rate": 0.00014572423233046386,
+      "loss": 2.152,
+      "step": 20
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 0.32099488413380073,
+      "learning_rate": 7.470666176083192e-05,
+      "loss": 1.8198,
+      "step": 30
+    },
+    {
+      "epoch": 2.4242424242424243,
+      "grad_norm": 0.2622629063842015,
+      "learning_rate": 1.660021821101222e-05,
+      "loss": 1.7642,
+      "step": 40
+    },
+    {
+      "epoch": 2.909090909090909,
+      "step": 48,
+      "total_flos": 88796276195328.0,
+      "train_loss": 2.5086214343706765,
+      "train_runtime": 154.8962,
+      "train_samples_per_second": 4.997,
+      "train_steps_per_second": 0.31
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 48,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 88796276195328.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}