Model save

README.md

@@ -0,0 +1,69 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+license: llama3
+tags:
+- trl
+- sft
+- generated_from_trainer
+model-index:
+- name: llama3-sudo
+  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. -->
+
+# llama3-sudo
+
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.0598
+
+## 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: 8
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 256
+- total_eval_batch_size: 16
+- 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 |
+|:-------------:|:------:|:----:|:---------------:|
+| 1.3252        | 0.9697 | 24   | 1.1693          |
+| 1.1352        | 1.9798 | 49   | 1.0709          |
+| 1.1558        | 2.9091 | 72   | 1.0598          |
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.44.0
+- Pytorch 2.1.2
+- Datasets 2.20.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.909090909090909,
+    "total_flos": 644347544469504.0,
+    "train_loss": 1.4444001482592688,
+    "train_runtime": 1573.8006,
+    "train_samples": 6321,
+    "train_samples_per_second": 12.049,
+    "train_steps_per_second": 0.046
+}

runs/Aug13_01-22-38_ip-172-31-10-237/events.out.tfevents.1723512170.ip-172-31-10-237.632570.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62e1296131d92859d57e8c03e820af57396fc15fc995acc8fd4c758bb0d85495
-size 9196
+oid sha256:476a7185484d3f0ce4bd728a827b00e92b53ffb34237a690341b87e78df58870
+size 9810

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.909090909090909,
+    "total_flos": 644347544469504.0,
+    "train_loss": 1.4444001482592688,
+    "train_runtime": 1573.8006,
+    "train_samples": 6321,
+    "train_samples_per_second": 12.049,
+    "train_steps_per_second": 0.046
+}

trainer_state.json

@@ -0,0 +1,171 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.909090909090909,
+  "eval_steps": 500,
+  "global_step": 72,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04040404040404041,
+      "grad_norm": 2.354057369065775,
+      "learning_rate": 2.5e-05,
+      "loss": 3.6821,
+      "step": 1
+    },
+    {
+      "epoch": 0.20202020202020202,
+      "grad_norm": 3.318817777883566,
+      "learning_rate": 0.000125,
+      "loss": 3.6411,
+      "step": 5
+    },
+    {
+      "epoch": 0.40404040404040403,
+      "grad_norm": 1.1567295340838406,
+      "learning_rate": 0.0001995184726672197,
+      "loss": 2.3671,
+      "step": 10
+    },
+    {
+      "epoch": 0.6060606060606061,
+      "grad_norm": 0.5651519215666836,
+      "learning_rate": 0.00019415440651830208,
+      "loss": 1.478,
+      "step": 15
+    },
+    {
+      "epoch": 0.8080808080808081,
+      "grad_norm": 0.4028700304940561,
+      "learning_rate": 0.00018314696123025454,
+      "loss": 1.3252,
+      "step": 20
+    },
+    {
+      "epoch": 0.9696969696969697,
+      "eval_loss": 1.1692708730697632,
+      "eval_runtime": 177.3352,
+      "eval_samples_per_second": 35.644,
+      "eval_steps_per_second": 2.233,
+      "step": 24
+    },
+    {
+      "epoch": 1.0101010101010102,
+      "grad_norm": 0.3079834505872443,
+      "learning_rate": 0.00016715589548470185,
+      "loss": 1.2706,
+      "step": 25
+    },
+    {
+      "epoch": 1.2121212121212122,
+      "grad_norm": 0.25648814038671125,
+      "learning_rate": 0.0001471396736825998,
+      "loss": 1.1881,
+      "step": 30
+    },
+    {
+      "epoch": 1.4141414141414141,
+      "grad_norm": 0.2008341133631863,
+      "learning_rate": 0.0001242980179903264,
+      "loss": 1.1514,
+      "step": 35
+    },
+    {
+      "epoch": 1.6161616161616161,
+      "grad_norm": 0.1814019139468598,
+      "learning_rate": 0.0001,
+      "loss": 1.1815,
+      "step": 40
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 0.18331867012606626,
+      "learning_rate": 7.570198200967362e-05,
+      "loss": 1.1352,
+      "step": 45
+    },
+    {
+      "epoch": 1.9797979797979797,
+      "eval_loss": 1.0709240436553955,
+      "eval_runtime": 175.2955,
+      "eval_samples_per_second": 36.059,
+      "eval_steps_per_second": 2.259,
+      "step": 49
+    },
+    {
+      "epoch": 2.0202020202020203,
+      "grad_norm": 0.2143290425477542,
+      "learning_rate": 5.286032631740023e-05,
+      "loss": 1.1283,
+      "step": 50
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 0.18532120253500695,
+      "learning_rate": 3.2844104515298155e-05,
+      "loss": 1.112,
+      "step": 55
+    },
+    {
+      "epoch": 2.4242424242424243,
+      "grad_norm": 0.19523394937897012,
+      "learning_rate": 1.6853038769745467e-05,
+      "loss": 1.1094,
+      "step": 60
+    },
+    {
+      "epoch": 2.6262626262626263,
+      "grad_norm": 0.19014746927385398,
+      "learning_rate": 5.8455934816979305e-06,
+      "loss": 1.1021,
+      "step": 65
+    },
+    {
+      "epoch": 2.8282828282828283,
+      "grad_norm": 0.1651175400452345,
+      "learning_rate": 4.815273327803182e-07,
+      "loss": 1.1558,
+      "step": 70
+    },
+    {
+      "epoch": 2.909090909090909,
+      "eval_loss": 1.0598390102386475,
+      "eval_runtime": 175.9252,
+      "eval_samples_per_second": 35.93,
+      "eval_steps_per_second": 2.251,
+      "step": 72
+    },
+    {
+      "epoch": 2.909090909090909,
+      "step": 72,
+      "total_flos": 644347544469504.0,
+      "train_loss": 1.4444001482592688,
+      "train_runtime": 1573.8006,
+      "train_samples_per_second": 12.049,
+      "train_steps_per_second": 0.046
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 72,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 25,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 644347544469504.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}