Model save

README.md

@@ -0,0 +1,70 @@
+---
+library_name: transformers
+license: gemma
+base_model: google/gemma-2-2b-it
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: gemma2b-b
+  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. -->
+
+# gemma2b-b
+
+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.
+It achieves the following results on the evaluation set:
+- Loss: 0.9842
+
+## 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: 2e-05
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 64
+- total_eval_batch_size: 8
+- 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
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 1.3036        | 0.9180 | 7    | 1.2136          |
+| 0.9604        | 1.9672 | 15   | 0.9997          |
+| 0.8292        | 2.7541 | 21   | 0.9842          |
+
+
+### Framework versions
+
+- Transformers 4.44.2
+- Pytorch 2.4.0+cu121
+- Datasets 3.0.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.7540983606557377,
+    "total_flos": 3971544514560.0,
+    "train_loss": 1.268857215132032,
+    "train_runtime": 396.9611,
+    "train_samples": 726,
+    "train_samples_per_second": 3.673,
+    "train_steps_per_second": 0.053
+}

generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": [
+    1,
+    107
+  ],
+  "pad_token_id": 0,
+  "transformers_version": "4.44.2"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.7540983606557377,
+    "total_flos": 3971544514560.0,
+    "train_loss": 1.268857215132032,
+    "train_runtime": 396.9611,
+    "train_samples": 726,
+    "train_samples_per_second": 3.673,
+    "train_steps_per_second": 0.053
+}

trainer_state.json

@@ -0,0 +1,213 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.7540983606557377,
+  "eval_steps": 500,
+  "global_step": 21,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.13114754098360656,
+      "grad_norm": 39.05130772271081,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 2.4376,
+      "step": 1
+    },
+    {
+      "epoch": 0.26229508196721313,
+      "grad_norm": 39.015973031999486,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 2.4339,
+      "step": 2
+    },
+    {
+      "epoch": 0.39344262295081966,
+      "grad_norm": 15.152312064538828,
+      "learning_rate": 2e-05,
+      "loss": 1.8888,
+      "step": 3
+    },
+    {
+      "epoch": 0.5245901639344263,
+      "grad_norm": 29.273723674546083,
+      "learning_rate": 1.9848077530122083e-05,
+      "loss": 1.8725,
+      "step": 4
+    },
+    {
+      "epoch": 0.6557377049180327,
+      "grad_norm": 11.667536919946288,
+      "learning_rate": 1.9396926207859085e-05,
+      "loss": 1.7797,
+      "step": 5
+    },
+    {
+      "epoch": 0.7868852459016393,
+      "grad_norm": 5.696635279505377,
+      "learning_rate": 1.866025403784439e-05,
+      "loss": 1.4526,
+      "step": 6
+    },
+    {
+      "epoch": 0.9180327868852459,
+      "grad_norm": 6.1474254223603655,
+      "learning_rate": 1.766044443118978e-05,
+      "loss": 1.3036,
+      "step": 7
+    },
+    {
+      "epoch": 0.9180327868852459,
+      "eval_loss": 1.2136186361312866,
+      "eval_runtime": 9.285,
+      "eval_samples_per_second": 20.678,
+      "eval_steps_per_second": 2.585,
+      "step": 7
+    },
+    {
+      "epoch": 1.0491803278688525,
+      "grad_norm": 3.8441152970886963,
+      "learning_rate": 1.6427876096865394e-05,
+      "loss": 1.193,
+      "step": 8
+    },
+    {
+      "epoch": 1.180327868852459,
+      "grad_norm": 3.216661811188678,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 1.1161,
+      "step": 9
+    },
+    {
+      "epoch": 1.3114754098360657,
+      "grad_norm": 2.783161962423741,
+      "learning_rate": 1.342020143325669e-05,
+      "loss": 1.0656,
+      "step": 10
+    },
+    {
+      "epoch": 1.4426229508196722,
+      "grad_norm": 2.6684579204452517,
+      "learning_rate": 1.1736481776669307e-05,
+      "loss": 1.0333,
+      "step": 11
+    },
+    {
+      "epoch": 1.5737704918032787,
+      "grad_norm": 2.1293181551342286,
+      "learning_rate": 1e-05,
+      "loss": 1.0045,
+      "step": 12
+    },
+    {
+      "epoch": 1.7049180327868854,
+      "grad_norm": 1.7349053966220533,
+      "learning_rate": 8.263518223330698e-06,
+      "loss": 0.971,
+      "step": 13
+    },
+    {
+      "epoch": 1.8360655737704918,
+      "grad_norm": 1.708605250889302,
+      "learning_rate": 6.579798566743314e-06,
+      "loss": 0.9746,
+      "step": 14
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "grad_norm": 1.565922544143404,
+      "learning_rate": 5.000000000000003e-06,
+      "loss": 0.9604,
+      "step": 15
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "eval_loss": 0.9996973872184753,
+      "eval_runtime": 9.0907,
+      "eval_samples_per_second": 21.121,
+      "eval_steps_per_second": 2.64,
+      "step": 15
+    },
+    {
+      "epoch": 2.098360655737705,
+      "grad_norm": 1.4703758003535075,
+      "learning_rate": 3.5721239031346067e-06,
+      "loss": 0.9006,
+      "step": 16
+    },
+    {
+      "epoch": 2.2295081967213113,
+      "grad_norm": 1.4546503498400736,
+      "learning_rate": 2.339555568810221e-06,
+      "loss": 0.8672,
+      "step": 17
+    },
+    {
+      "epoch": 2.360655737704918,
+      "grad_norm": 1.333908386274604,
+      "learning_rate": 1.339745962155613e-06,
+      "loss": 0.8603,
+      "step": 18
+    },
+    {
+      "epoch": 2.4918032786885247,
+      "grad_norm": 1.2309918853415422,
+      "learning_rate": 6.030737921409169e-07,
+      "loss": 0.8533,
+      "step": 19
+    },
+    {
+      "epoch": 2.6229508196721314,
+      "grad_norm": 1.187570177680188,
+      "learning_rate": 1.519224698779198e-07,
+      "loss": 0.8482,
+      "step": 20
+    },
+    {
+      "epoch": 2.7540983606557377,
+      "grad_norm": 1.1796688649183953,
+      "learning_rate": 0.0,
+      "loss": 0.8292,
+      "step": 21
+    },
+    {
+      "epoch": 2.7540983606557377,
+      "eval_loss": 0.9841778874397278,
+      "eval_runtime": 9.0575,
+      "eval_samples_per_second": 21.198,
+      "eval_steps_per_second": 2.65,
+      "step": 21
+    },
+    {
+      "epoch": 2.7540983606557377,
+      "step": 21,
+      "total_flos": 3971544514560.0,
+      "train_loss": 1.268857215132032,
+      "train_runtime": 396.9611,
+      "train_samples_per_second": 3.673,
+      "train_steps_per_second": 0.053
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 21,
+  "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": 3971544514560.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}