Model save

README.md

@@ -0,0 +1,63 @@
+---
+library_name: transformers
+license: gemma
+base_model: google/gemma-2-9b-it
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: gemma9b-sft-m
+  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. -->
+
+# gemma9b-sft-m
+
+This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-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: 2e-05
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 128
+- 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: 5
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.45.2
+- Pytorch 2.1.2+cu121
+- Datasets 3.0.1
+- Tokenizers 0.20.0

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.324324324324325,
+    "total_flos": 2.618567468158812e+17,
+    "train_loss": 0.8209461569786072,
+    "train_runtime": 1075.8042,
+    "train_samples": 7473,
+    "train_samples_per_second": 2.751,
+    "train_steps_per_second": 0.019
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.45.2"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.324324324324325,
+    "total_flos": 2.618567468158812e+17,
+    "train_loss": 0.8209461569786072,
+    "train_runtime": 1075.8042,
+    "train_samples": 7473,
+    "train_samples_per_second": 2.751,
+    "train_steps_per_second": 0.019
+}

trainer_state.json

@@ -0,0 +1,182 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.324324324324325,
+  "eval_steps": 500,
+  "global_step": 20,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.21621621621621623,
+      "grad_norm": 16.411664962768555,
+      "learning_rate": 1e-05,
+      "loss": 1.2709,
+      "step": 1
+    },
+    {
+      "epoch": 0.43243243243243246,
+      "grad_norm": 15.362375259399414,
+      "learning_rate": 2e-05,
+      "loss": 1.2592,
+      "step": 2
+    },
+    {
+      "epoch": 0.6486486486486487,
+      "grad_norm": 9.707308769226074,
+      "learning_rate": 1.9848077530122083e-05,
+      "loss": 1.0992,
+      "step": 3
+    },
+    {
+      "epoch": 0.8648648648648649,
+      "grad_norm": 21.78615951538086,
+      "learning_rate": 1.9396926207859085e-05,
+      "loss": 1.1438,
+      "step": 4
+    },
+    {
+      "epoch": 1.0810810810810811,
+      "grad_norm": 7.769772529602051,
+      "learning_rate": 1.866025403784439e-05,
+      "loss": 0.9555,
+      "step": 5
+    },
+    {
+      "epoch": 1.2972972972972974,
+      "grad_norm": 3.812919855117798,
+      "learning_rate": 1.766044443118978e-05,
+      "loss": 0.8601,
+      "step": 6
+    },
+    {
+      "epoch": 1.5135135135135136,
+      "grad_norm": 10.630844116210938,
+      "learning_rate": 1.6427876096865394e-05,
+      "loss": 0.9599,
+      "step": 7
+    },
+    {
+      "epoch": 1.7297297297297298,
+      "grad_norm": 3.042130708694458,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.8358,
+      "step": 8
+    },
+    {
+      "epoch": 1.945945945945946,
+      "grad_norm": 5.21888542175293,
+      "learning_rate": 1.342020143325669e-05,
+      "loss": 0.8226,
+      "step": 9
+    },
+    {
+      "epoch": 2.1621621621621623,
+      "grad_norm": 1.5614551305770874,
+      "learning_rate": 1.1736481776669307e-05,
+      "loss": 0.7464,
+      "step": 10
+    },
+    {
+      "epoch": 2.3783783783783785,
+      "grad_norm": 1.4593987464904785,
+      "learning_rate": 1e-05,
+      "loss": 0.7227,
+      "step": 11
+    },
+    {
+      "epoch": 2.5945945945945947,
+      "grad_norm": 1.3283805847167969,
+      "learning_rate": 8.263518223330698e-06,
+      "loss": 0.7012,
+      "step": 12
+    },
+    {
+      "epoch": 2.810810810810811,
+      "grad_norm": 1.1728136539459229,
+      "learning_rate": 6.579798566743314e-06,
+      "loss": 0.6956,
+      "step": 13
+    },
+    {
+      "epoch": 3.027027027027027,
+      "grad_norm": 0.7944673299789429,
+      "learning_rate": 5.000000000000003e-06,
+      "loss": 0.6836,
+      "step": 14
+    },
+    {
+      "epoch": 3.2432432432432434,
+      "grad_norm": 0.8301619291305542,
+      "learning_rate": 3.5721239031346067e-06,
+      "loss": 0.6346,
+      "step": 15
+    },
+    {
+      "epoch": 3.4594594594594597,
+      "grad_norm": 0.7796211242675781,
+      "learning_rate": 2.339555568810221e-06,
+      "loss": 0.6218,
+      "step": 16
+    },
+    {
+      "epoch": 3.6756756756756754,
+      "grad_norm": 0.6915468573570251,
+      "learning_rate": 1.339745962155613e-06,
+      "loss": 0.6058,
+      "step": 17
+    },
+    {
+      "epoch": 3.891891891891892,
+      "grad_norm": 0.6428016424179077,
+      "learning_rate": 6.030737921409169e-07,
+      "loss": 0.6106,
+      "step": 18
+    },
+    {
+      "epoch": 4.108108108108108,
+      "grad_norm": 0.6951332688331604,
+      "learning_rate": 1.519224698779198e-07,
+      "loss": 0.5922,
+      "step": 19
+    },
+    {
+      "epoch": 4.324324324324325,
+      "grad_norm": 0.6421741247177124,
+      "learning_rate": 0.0,
+      "loss": 0.5973,
+      "step": 20
+    },
+    {
+      "epoch": 4.324324324324325,
+      "step": 20,
+      "total_flos": 2.618567468158812e+17,
+      "train_loss": 0.8209461569786072,
+      "train_runtime": 1075.8042,
+      "train_samples_per_second": 2.751,
+      "train_steps_per_second": 0.019
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 20,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 999999,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.618567468158812e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}