Model save

README.md

@@ -0,0 +1,70 @@
+---
+license: gemma
+library_name: peft
+tags:
+- trl
+- sft
+- generated_from_trainer
+base_model: google/gemma-7b
+datasets:
+- generator
+model-index:
+- name: gemma7b-summarize-gemini1.5flash-30k
+  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. -->
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/chansung18/huggingface/runs/6tjy33fd)
+# gemma7b-summarize-gemini1.5flash-30k
+
+This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.2895
+
+## 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: 0.0002
+- train_batch_size: 4
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- total_eval_batch_size: 4
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 0.8678        | 0.9956 | 114  | 2.2895          |
+
+
+### Framework versions
+
+- PEFT 0.11.1
+- Transformers 4.41.0
+- Pytorch 2.3.0+cu121
+- Datasets 2.19.1
+- Tokenizers 0.19.1

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b5efbefe5ecdf4e4c499137fbf6a73833100c452fee84db26f0a48501b8a0d5
+oid sha256:7d3bac2018fb39764f96cf64f3e1b1aa8334f137d5effe5453b4439958689932
 size 50056096

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9956331877729258,
+    "total_flos": 1.742915407875932e+17,
+    "train_loss": 5.99931854532476,
+    "train_runtime": 898.0288,
+    "train_samples": 30386,
+    "train_samples_per_second": 2.039,
+    "train_steps_per_second": 0.127
+}

runs/May19_11-18-25_deep-diver-main-lucky-bat-1-0-0/events.out.tfevents.1716132059.deep-diver-main-lucky-bat-1-0-0.385.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d49632e1c5aeca5cf7d1dc4f8f3eeb2f3f83be0f3721c5c8eaa86fd738fc167b
-size 9971
+oid sha256:a3e783394fb3d1cd74be49ffe8d46507f8ad4313d9e5a776e9e5aaf81306f3b9
+size 10999

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.9956331877729258,
+    "total_flos": 1.742915407875932e+17,
+    "train_loss": 5.99931854532476,
+    "train_runtime": 898.0288,
+    "train_samples": 30386,
+    "train_samples_per_second": 2.039,
+    "train_steps_per_second": 0.127
+}

trainer_state.json

@@ -0,0 +1,211 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9956331877729258,
+  "eval_steps": 500,
+  "global_step": 114,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008733624454148471,
+      "grad_norm": 648.0,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 56.9969,
+      "step": 1
+    },
+    {
+      "epoch": 0.043668122270742356,
+      "grad_norm": 394.0,
+      "learning_rate": 8.333333333333334e-05,
+      "loss": 51.7017,
+      "step": 5
+    },
+    {
+      "epoch": 0.08733624454148471,
+      "grad_norm": 30.25,
+      "learning_rate": 0.0001666666666666667,
+      "loss": 22.1529,
+      "step": 10
+    },
+    {
+      "epoch": 0.13100436681222707,
+      "grad_norm": 8.8125,
+      "learning_rate": 0.00019957341762950344,
+      "loss": 17.6564,
+      "step": 15
+    },
+    {
+      "epoch": 0.17467248908296942,
+      "grad_norm": 23.0,
+      "learning_rate": 0.00019697969360350098,
+      "loss": 15.7963,
+      "step": 20
+    },
+    {
+      "epoch": 0.2183406113537118,
+      "grad_norm": 108.5,
+      "learning_rate": 0.0001920905517944954,
+      "loss": 8.7143,
+      "step": 25
+    },
+    {
+      "epoch": 0.26200873362445415,
+      "grad_norm": 7.9375,
+      "learning_rate": 0.00018502171357296144,
+      "loss": 2.0425,
+      "step": 30
+    },
+    {
+      "epoch": 0.3056768558951965,
+      "grad_norm": 22.5,
+      "learning_rate": 0.00017594049166547073,
+      "loss": 1.6476,
+      "step": 35
+    },
+    {
+      "epoch": 0.34934497816593885,
+      "grad_norm": 1.4765625,
+      "learning_rate": 0.0001650618300204242,
+      "loss": 1.3805,
+      "step": 40
+    },
+    {
+      "epoch": 0.3930131004366812,
+      "grad_norm": 2.109375,
+      "learning_rate": 0.0001526432162877356,
+      "loss": 1.2737,
+      "step": 45
+    },
+    {
+      "epoch": 0.4366812227074236,
+      "grad_norm": 3.0625,
+      "learning_rate": 0.00013897858732926793,
+      "loss": 1.2169,
+      "step": 50
+    },
+    {
+      "epoch": 0.48034934497816595,
+      "grad_norm": 1.7734375,
+      "learning_rate": 0.00012439137201083773,
+      "loss": 1.1317,
+      "step": 55
+    },
+    {
+      "epoch": 0.5240174672489083,
+      "grad_norm": 1.515625,
+      "learning_rate": 0.00010922683594633021,
+      "loss": 1.0869,
+      "step": 60
+    },
+    {
+      "epoch": 0.5676855895196506,
+      "grad_norm": 1.7578125,
+      "learning_rate": 9.384390938660572e-05,
+      "loss": 1.0091,
+      "step": 65
+    },
+    {
+      "epoch": 0.611353711790393,
+      "grad_norm": 2.5625,
+      "learning_rate": 7.860669167935028e-05,
+      "loss": 1.0185,
+      "step": 70
+    },
+    {
+      "epoch": 0.6550218340611353,
+      "grad_norm": 1.28125,
+      "learning_rate": 6.387583338128471e-05,
+      "loss": 0.9505,
+      "step": 75
+    },
+    {
+      "epoch": 0.6986899563318777,
+      "grad_norm": 1.6796875,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 0.9331,
+      "step": 80
+    },
+    {
+      "epoch": 0.74235807860262,
+      "grad_norm": 1.7734375,
+      "learning_rate": 3.7307619410589376e-05,
+      "loss": 0.9083,
+      "step": 85
+    },
+    {
+      "epoch": 0.7860262008733624,
+      "grad_norm": 0.9921875,
+      "learning_rate": 2.6099108277934103e-05,
+      "loss": 0.884,
+      "step": 90
+    },
+    {
+      "epoch": 0.8296943231441049,
+      "grad_norm": 1.328125,
+      "learning_rate": 1.663976147788806e-05,
+      "loss": 0.9173,
+      "step": 95
+    },
+    {
+      "epoch": 0.8733624454148472,
+      "grad_norm": 0.85546875,
+      "learning_rate": 9.153472818047625e-06,
+      "loss": 0.8598,
+      "step": 100
+    },
+    {
+      "epoch": 0.9170305676855895,
+      "grad_norm": 0.69140625,
+      "learning_rate": 3.817435682718096e-06,
+      "loss": 0.8639,
+      "step": 105
+    },
+    {
+      "epoch": 0.9606986899563319,
+      "grad_norm": 0.71875,
+      "learning_rate": 7.579490328064265e-07,
+      "loss": 0.8678,
+      "step": 110
+    },
+    {
+      "epoch": 0.9956331877729258,
+      "eval_loss": 2.2895236015319824,
+      "eval_runtime": 1.0009,
+      "eval_samples_per_second": 4.996,
+      "eval_steps_per_second": 1.998,
+      "step": 114
+    },
+    {
+      "epoch": 0.9956331877729258,
+      "step": 114,
+      "total_flos": 1.742915407875932e+17,
+      "train_loss": 5.99931854532476,
+      "train_runtime": 898.0288,
+      "train_samples_per_second": 2.039,
+      "train_steps_per_second": 0.127
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 114,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.742915407875932e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}