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-8
+  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-8
+
+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: 0.0276
+
+## 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: 4
+- total_train_batch_size: 32
+- total_eval_batch_size: 32
+- 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 |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.0201        | 1.0   | 4    | 0.8704          |
+| 0.9676        | 2.0   | 8    | 0.6483          |
+| 0.7053        | 3.0   | 12   | 0.4601          |
+| 0.4784        | 4.0   | 16   | 0.2716          |
+| 0.2487        | 5.0   | 20   | 0.1375          |
+| 0.2487        | 6.0   | 24   | 0.0688          |
+| 0.1048        | 7.0   | 28   | 0.0426          |
+| 0.0523        | 8.0   | 32   | 0.0332          |
+| 0.0354        | 9.0   | 36   | 0.0287          |
+| 0.0285        | 10.0  | 40   | 0.0276          |
+
+
+### 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": 8375186227200.0,
+    "train_loss": 0.3289382763206959,
+    "train_runtime": 602.9677,
+    "train_samples": 2000,
+    "train_samples_per_second": 1.775,
+    "train_steps_per_second": 0.066
+}

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": 8375186227200.0,
+    "train_loss": 0.3289382763206959,
+    "train_runtime": 602.9677,
+    "train_samples": 2000,
+    "train_samples_per_second": 1.775,
+    "train_steps_per_second": 0.066
+}

trainer_state.json

@@ -0,0 +1,185 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 40,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.25,
+      "grad_norm": 36.86391077634519,
+      "learning_rate": 5e-06,
+      "loss": 1.0201,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.8703737258911133,
+      "eval_runtime": 11.8087,
+      "eval_samples_per_second": 9.061,
+      "eval_steps_per_second": 0.339,
+      "step": 4
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 6.912887702306342,
+      "learning_rate": 9.847001329696653e-06,
+      "loss": 0.9676,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6482987403869629,
+      "eval_runtime": 10.5317,
+      "eval_samples_per_second": 10.16,
+      "eval_steps_per_second": 0.38,
+      "step": 8
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 3.426473054986459,
+      "learning_rate": 8.94570254698197e-06,
+      "loss": 0.7053,
+      "step": 10
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.4600563645362854,
+      "eval_runtime": 10.5697,
+      "eval_samples_per_second": 10.123,
+      "eval_steps_per_second": 0.378,
+      "step": 12
+    },
+    {
+      "epoch": 3.75,
+      "grad_norm": 3.7980492702489292,
+      "learning_rate": 7.379736965185369e-06,
+      "loss": 0.4784,
+      "step": 15
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.2715964913368225,
+      "eval_runtime": 10.622,
+      "eval_samples_per_second": 10.073,
+      "eval_steps_per_second": 0.377,
+      "step": 16
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 5.426565998488602,
+      "learning_rate": 5.412896727361663e-06,
+      "loss": 0.2487,
+      "step": 20
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.13748349249362946,
+      "eval_runtime": 10.5895,
+      "eval_samples_per_second": 10.104,
+      "eval_steps_per_second": 0.378,
+      "step": 20
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.0687556117773056,
+      "eval_runtime": 10.6498,
+      "eval_samples_per_second": 10.047,
+      "eval_steps_per_second": 0.376,
+      "step": 24
+    },
+    {
+      "epoch": 6.25,
+      "grad_norm": 2.4213215467933678,
+      "learning_rate": 3.3765026539765832e-06,
+      "loss": 0.1048,
+      "step": 25
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.04261770844459534,
+      "eval_runtime": 10.5949,
+      "eval_samples_per_second": 10.099,
+      "eval_steps_per_second": 0.378,
+      "step": 28
+    },
+    {
+      "epoch": 7.5,
+      "grad_norm": 1.4054408580810511,
+      "learning_rate": 1.6135921418712959e-06,
+      "loss": 0.0523,
+      "step": 30
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.03317110612988472,
+      "eval_runtime": 10.6578,
+      "eval_samples_per_second": 10.04,
+      "eval_steps_per_second": 0.375,
+      "step": 32
+    },
+    {
+      "epoch": 8.75,
+      "grad_norm": 0.8161983996171017,
+      "learning_rate": 4.211333667247125e-07,
+      "loss": 0.0354,
+      "step": 35
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 0.028676806017756462,
+      "eval_runtime": 10.5748,
+      "eval_samples_per_second": 10.118,
+      "eval_steps_per_second": 0.378,
+      "step": 36
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.6106701044219125,
+      "learning_rate": 0.0,
+      "loss": 0.0285,
+      "step": 40
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 0.027601005509495735,
+      "eval_runtime": 10.3944,
+      "eval_samples_per_second": 10.294,
+      "eval_steps_per_second": 0.385,
+      "step": 40
+    },
+    {
+      "epoch": 10.0,
+      "step": 40,
+      "total_flos": 8375186227200.0,
+      "train_loss": 0.3289382763206959,
+      "train_runtime": 602.9677,
+      "train_samples_per_second": 1.775,
+      "train_steps_per_second": 0.066
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 40,
+  "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": 8375186227200.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}