Model save

README.md

@@ -0,0 +1,69 @@
+---
+library_name: peft
+license: llama3
+base_model: meta-llama/Meta-Llama-3-8B
+tags:
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- generator
+model-index:
+- name: llama3-8b-pissa-classification-11-v1
+  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. -->
+
+# llama3-8b-pissa-classification-11-v1
+
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.2708
+
+## 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: 16
+- eval_batch_size: 16
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 256
+- total_eval_batch_size: 128
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 2.1904        | 1.0   | 61   | 2.2708          |
+
+
+### Framework versions
+
+- PEFT 0.13.2
+- Transformers 4.46.2
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 3.60192752912171e+17,
+    "train_loss": 2.376973159977647,
+    "train_runtime": 294.1531,
+    "train_samples": 92634,
+    "train_samples_per_second": 52.704,
+    "train_steps_per_second": 0.207
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 3.60192752912171e+17,
+    "train_loss": 2.376973159977647,
+    "train_runtime": 294.1531,
+    "train_samples": 92634,
+    "train_samples_per_second": 52.704,
+    "train_steps_per_second": 0.207
+}

trainer_state.json

@@ -0,0 +1,141 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 61,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01639344262295082,
+      "grad_norm": 163.37210083007812,
+      "learning_rate": 2.857142857142857e-05,
+      "loss": 3.2542,
+      "step": 1
+    },
+    {
+      "epoch": 0.08196721311475409,
+      "grad_norm": 2.698173999786377,
+      "learning_rate": 0.00014285714285714287,
+      "loss": 3.0769,
+      "step": 5
+    },
+    {
+      "epoch": 0.16393442622950818,
+      "grad_norm": 2.6624538898468018,
+      "learning_rate": 0.00019848077530122083,
+      "loss": 2.764,
+      "step": 10
+    },
+    {
+      "epoch": 0.2459016393442623,
+      "grad_norm": 0.9797217845916748,
+      "learning_rate": 0.00018936326403234125,
+      "loss": 2.5387,
+      "step": 15
+    },
+    {
+      "epoch": 0.32786885245901637,
+      "grad_norm": 1.1679744720458984,
+      "learning_rate": 0.00017273736415730488,
+      "loss": 2.3838,
+      "step": 20
+    },
+    {
+      "epoch": 0.4098360655737705,
+      "grad_norm": 0.8691921830177307,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 2.3067,
+      "step": 25
+    },
+    {
+      "epoch": 0.4918032786885246,
+      "grad_norm": 0.6480523943901062,
+      "learning_rate": 0.00012306158707424403,
+      "loss": 2.2564,
+      "step": 30
+    },
+    {
+      "epoch": 0.5737704918032787,
+      "grad_norm": 0.3995227813720703,
+      "learning_rate": 9.418551710895243e-05,
+      "loss": 2.2165,
+      "step": 35
+    },
+    {
+      "epoch": 0.6557377049180327,
+      "grad_norm": 0.2950589060783386,
+      "learning_rate": 6.579798566743314e-05,
+      "loss": 2.2065,
+      "step": 40
+    },
+    {
+      "epoch": 0.7377049180327869,
+      "grad_norm": 0.33247917890548706,
+      "learning_rate": 4.028414082972141e-05,
+      "loss": 2.1951,
+      "step": 45
+    },
+    {
+      "epoch": 0.819672131147541,
+      "grad_norm": 0.520527184009552,
+      "learning_rate": 1.9787680724495617e-05,
+      "loss": 2.1865,
+      "step": 50
+    },
+    {
+      "epoch": 0.9016393442622951,
+      "grad_norm": 0.3367004692554474,
+      "learning_rate": 6.030737921409169e-06,
+      "loss": 2.1967,
+      "step": 55
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "grad_norm": 0.3582340180873871,
+      "learning_rate": 1.6918417287318245e-07,
+      "loss": 2.1904,
+      "step": 60
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.2707738876342773,
+      "eval_runtime": 0.8017,
+      "eval_samples_per_second": 4.989,
+      "eval_steps_per_second": 1.247,
+      "step": 61
+    },
+    {
+      "epoch": 1.0,
+      "step": 61,
+      "total_flos": 3.60192752912171e+17,
+      "train_loss": 2.376973159977647,
+      "train_runtime": 294.1531,
+      "train_samples_per_second": 52.704,
+      "train_steps_per_second": 0.207
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 61,
+  "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": 3.60192752912171e+17,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}