Model save

README.md

@@ -0,0 +1,57 @@
+---
+base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
+library_name: transformers
+model_name: llama-sft-qlora
+tags:
+- generated_from_trainer
+- trl
+- sft
+licence: license
+---
+
+# Model Card for llama-sft-qlora
+
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="fenguhao/llama-sft-qlora", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.12.1
+- Transformers: 4.46.3
+- Pytorch: 2.5.1
+- Datasets: 3.1.0
+- Tokenizers: 0.20.3
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 5.19321114598441e+16,
+    "train_loss": 0.29687737567084177,
+    "train_runtime": 332.0278,
+    "train_samples": 9000,
+    "train_samples_per_second": 1.678,
+    "train_steps_per_second": 0.105
+}

runs/Nov29_08-03-41_RLHF000/events.out.tfevents.1732867429.RLHF000.552686.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13ae527b1e18dae6ac206c66caead9bab20da37ee46dcb84874b22ff7c7dd318
-size 8549
+oid sha256:5f99edd6daae5bc4bd9da8acd8638327cf11b39f5443fabcedb54d179e9b2680
+size 9163

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 5.19321114598441e+16,
+    "train_loss": 0.29687737567084177,
+    "train_runtime": 332.0278,
+    "train_samples": 9000,
+    "train_samples_per_second": 1.678,
+    "train_steps_per_second": 0.105
+}

trainer_state.json

@@ -0,0 +1,106 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 35,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02857142857142857,
+      "grad_norm": 1.1770261526107788,
+      "learning_rate": 5e-05,
+      "loss": 0.781,
+      "step": 1
+    },
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 0.7865971326828003,
+      "learning_rate": 0.00019948693233918952,
+      "loss": 0.6954,
+      "step": 5
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 0.3400590717792511,
+      "learning_rate": 0.00018207634412072764,
+      "loss": 0.4525,
+      "step": 10
+    },
+    {
+      "epoch": 0.42857142857142855,
+      "grad_norm": 0.24245862662792206,
+      "learning_rate": 0.00014403941515576344,
+      "loss": 0.306,
+      "step": 15
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.23055194318294525,
+      "learning_rate": 9.493508311612874e-05,
+      "loss": 0.2153,
+      "step": 20
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.25304117798805237,
+      "learning_rate": 4.710359896730379e-05,
+      "loss": 0.1602,
+      "step": 25
+    },
+    {
+      "epoch": 0.8571428571428571,
+      "grad_norm": 0.2968534231185913,
+      "learning_rate": 1.2565338385541792e-05,
+      "loss": 0.1206,
+      "step": 30
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.11835439503192902,
+      "learning_rate": 0.0,
+      "loss": 0.111,
+      "step": 35
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.11065573990345001,
+      "eval_runtime": 10.7768,
+      "eval_samples_per_second": 5.66,
+      "eval_steps_per_second": 0.371,
+      "step": 35
+    },
+    {
+      "epoch": 1.0,
+      "step": 35,
+      "total_flos": 5.19321114598441e+16,
+      "train_loss": 0.29687737567084177,
+      "train_runtime": 332.0278,
+      "train_samples_per_second": 1.678,
+      "train_steps_per_second": 0.105
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 35,
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.19321114598441e+16,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}