Model save

README.md

@@ -0,0 +1,70 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: mistralai/Mistral-Nemo-Base-2407
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: Mistral-Nemo-Instruct-MCAI-SFT-3Epochs
+  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. -->
+
+# Mistral-Nemo-Instruct-MCAI-SFT-3Epochs
+
+This model is a fine-tuned version of [mistralai/Mistral-Nemo-Base-2407](https://huggingface.co/mistralai/Mistral-Nemo-Base-2407) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.2052
+
+## 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: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 256
+- 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: 0.1
+- num_epochs: 3
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 1.9196        | 0.9811 | 13   | 1.6468          |
+| 1.2211        | 1.9623 | 26   | 1.2391          |
+| 1.0105        | 2.9434 | 39   | 1.2052          |
+
+
+### Framework versions
+
+- Transformers 4.45.1
+- Pytorch 2.4.1+cu121
+- Datasets 3.0.1
+- Tokenizers 0.20.0

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.9433962264150946,
+    "total_flos": 50553107251200.0,
+    "train_loss": 1.389819472264021,
+    "train_runtime": 1921.6463,
+    "train_samples": 19817,
+    "train_samples_per_second": 5.202,
+    "train_steps_per_second": 0.02
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.45.1"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.9433962264150946,
+    "total_flos": 50553107251200.0,
+    "train_loss": 1.389819472264021,
+    "train_runtime": 1921.6463,
+    "train_samples": 19817,
+    "train_samples_per_second": 5.202,
+    "train_steps_per_second": 0.02
+}

trainer_state.json

@@ -0,0 +1,122 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9433962264150946,
+  "eval_steps": 500,
+  "global_step": 39,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07547169811320754,
+      "grad_norm": 9.254759322777849e+19,
+      "learning_rate": 5e-06,
+      "loss": 1.8775,
+      "step": 1
+    },
+    {
+      "epoch": 0.37735849056603776,
+      "grad_norm": 52.04201026441062,
+      "learning_rate": 1.9959742939952393e-05,
+      "loss": 1.8165,
+      "step": 5
+    },
+    {
+      "epoch": 0.7547169811320755,
+      "grad_norm": 274.3938636993834,
+      "learning_rate": 1.8584487936018663e-05,
+      "loss": 1.9196,
+      "step": 10
+    },
+    {
+      "epoch": 0.9811320754716981,
+      "eval_loss": 1.6468288898468018,
+      "eval_runtime": 18.2105,
+      "eval_samples_per_second": 20.812,
+      "eval_steps_per_second": 0.329,
+      "step": 13
+    },
+    {
+      "epoch": 1.1320754716981132,
+      "grad_norm": 7.996669743347857,
+      "learning_rate": 1.5508969814521026e-05,
+      "loss": 1.6376,
+      "step": 15
+    },
+    {
+      "epoch": 1.509433962264151,
+      "grad_norm": 3.3866631549093404,
+      "learning_rate": 1.1342332658176556e-05,
+      "loss": 1.3432,
+      "step": 20
+    },
+    {
+      "epoch": 1.8867924528301887,
+      "grad_norm": 1.6559910834673521,
+      "learning_rate": 6.909830056250527e-06,
+      "loss": 1.2211,
+      "step": 25
+    },
+    {
+      "epoch": 1.9622641509433962,
+      "eval_loss": 1.2391045093536377,
+      "eval_runtime": 17.9683,
+      "eval_samples_per_second": 21.093,
+      "eval_steps_per_second": 0.334,
+      "step": 26
+    },
+    {
+      "epoch": 2.2641509433962264,
+      "grad_norm": 1.2242685027552433,
+      "learning_rate": 3.089373510131354e-06,
+      "loss": 1.0803,
+      "step": 30
+    },
+    {
+      "epoch": 2.641509433962264,
+      "grad_norm": 1.0045856779786766,
+      "learning_rate": 6.37651293602628e-07,
+      "loss": 1.0105,
+      "step": 35
+    },
+    {
+      "epoch": 2.9433962264150946,
+      "eval_loss": 1.20521080493927,
+      "eval_runtime": 17.9079,
+      "eval_samples_per_second": 21.164,
+      "eval_steps_per_second": 0.335,
+      "step": 39
+    },
+    {
+      "epoch": 2.9433962264150946,
+      "step": 39,
+      "total_flos": 50553107251200.0,
+      "train_loss": 1.389819472264021,
+      "train_runtime": 1921.6463,
+      "train_samples_per_second": 5.202,
+      "train_steps_per_second": 0.02
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 39,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "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": 50553107251200.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}