Model save

README.md

@@ -0,0 +1,76 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: alignment-handbook/zephyr-7b-sft-full
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: zephyr-7b-math-case-6
+  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. -->
+
+# zephyr-7b-math-case-6
+
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0198
+
+## 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: 1e-05
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 64
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.03
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.8827        | 1.0   | 5    | 0.7429          |
+| 0.6709        | 2.0   | 10   | 0.5531          |
+| 0.5071        | 3.0   | 15   | 0.4035          |
+| 0.3519        | 4.0   | 20   | 0.2455          |
+| 0.2036        | 5.0   | 25   | 0.1302          |
+| 0.1035        | 6.0   | 30   | 0.0602          |
+| 0.0527        | 7.0   | 35   | 0.0356          |
+| 0.0321        | 8.0   | 40   | 0.0249          |
+| 0.0236        | 9.0   | 45   | 0.0206          |
+| 0.0202        | 10.0  | 50   | 0.0198          |
+
+
+### Framework versions
+
+- Transformers 4.44.2
+- Pytorch 2.4.1+cu121
+- Datasets 3.0.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 20937965568000.0,
+    "train_loss": 0.28585494235157966,
+    "train_runtime": 720.2074,
+    "train_samples": 4000,
+    "train_samples_per_second": 3.652,
+    "train_steps_per_second": 0.069
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.44.2"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 20937965568000.0,
+    "train_loss": 0.28585494235157966,
+    "train_runtime": 720.2074,
+    "train_samples": 4000,
+    "train_samples_per_second": 3.652,
+    "train_steps_per_second": 0.069
+}

trainer_state.json

@@ -0,0 +1,199 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2,
+      "grad_norm": 24.890761216715067,
+      "learning_rate": 5e-06,
+      "loss": 0.9347,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 6.27322744929223,
+      "learning_rate": 9.903926402016153e-06,
+      "loss": 0.8827,
+      "step": 5
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.7428836822509766,
+      "eval_runtime": 13.5132,
+      "eval_samples_per_second": 19.463,
+      "eval_steps_per_second": 0.37,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.4763297090491645,
+      "learning_rate": 9.330127018922195e-06,
+      "loss": 0.6709,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.5530853867530823,
+      "eval_runtime": 13.3663,
+      "eval_samples_per_second": 19.676,
+      "eval_steps_per_second": 0.374,
+      "step": 10
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.9858549062244277,
+      "learning_rate": 8.296729075500345e-06,
+      "loss": 0.5071,
+      "step": 15
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.4035380482673645,
+      "eval_runtime": 13.3538,
+      "eval_samples_per_second": 19.695,
+      "eval_steps_per_second": 0.374,
+      "step": 15
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.012038571620908,
+      "learning_rate": 6.913417161825449e-06,
+      "loss": 0.3519,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.24551308155059814,
+      "eval_runtime": 13.3982,
+      "eval_samples_per_second": 19.63,
+      "eval_steps_per_second": 0.373,
+      "step": 20
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.9052862275719975,
+      "learning_rate": 5.327015646150716e-06,
+      "loss": 0.2036,
+      "step": 25
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.1301579624414444,
+      "eval_runtime": 13.3478,
+      "eval_samples_per_second": 19.704,
+      "eval_steps_per_second": 0.375,
+      "step": 25
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.7865230440629105,
+      "learning_rate": 3.705904774487396e-06,
+      "loss": 0.1035,
+      "step": 30
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.060157448053359985,
+      "eval_runtime": 13.3892,
+      "eval_samples_per_second": 19.643,
+      "eval_steps_per_second": 0.373,
+      "step": 30
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.798134146722217,
+      "learning_rate": 2.2221488349019903e-06,
+      "loss": 0.0527,
+      "step": 35
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.03557642176747322,
+      "eval_runtime": 13.3867,
+      "eval_samples_per_second": 19.646,
+      "eval_steps_per_second": 0.374,
+      "step": 35
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.2741605399529363,
+      "learning_rate": 1.0332332985438248e-06,
+      "loss": 0.0321,
+      "step": 40
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.024869758635759354,
+      "eval_runtime": 13.3698,
+      "eval_samples_per_second": 19.671,
+      "eval_steps_per_second": 0.374,
+      "step": 40
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.6581840143146274,
+      "learning_rate": 2.653493525244721e-07,
+      "loss": 0.0236,
+      "step": 45
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 0.02064005844295025,
+      "eval_runtime": 13.4054,
+      "eval_samples_per_second": 19.619,
+      "eval_steps_per_second": 0.373,
+      "step": 45
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.4976189909335096,
+      "learning_rate": 0.0,
+      "loss": 0.0202,
+      "step": 50
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 0.019776934757828712,
+      "eval_runtime": 13.3303,
+      "eval_samples_per_second": 19.729,
+      "eval_steps_per_second": 0.375,
+      "step": 50
+    },
+    {
+      "epoch": 10.0,
+      "step": 50,
+      "total_flos": 20937965568000.0,
+      "train_loss": 0.28585494235157966,
+      "train_runtime": 720.2074,
+      "train_samples_per_second": 3.652,
+      "train_steps_per_second": 0.069
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 50,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 20937965568000.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}