Model save

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,78 @@
+---
+license: gemma
+library_name: peft
+tags:
+- trl
+- sft
+- generated_from_trainer
+base_model: google/gemma-2b
+datasets:
+- generator
+model-index:
+- name: gemma2b-summarize-gemini1_5flash-4k
+  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. -->
+
+# gemma2b-summarize-gemini1_5flash-4k
+
+This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.5520
+
+## 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: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 2.9598        | 1.0   | 7    | 3.0819          |
+| 2.584         | 2.0   | 14   | 2.7799          |
+| 1.8743        | 3.0   | 21   | 2.6390          |
+| 1.6598        | 4.0   | 28   | 2.6231          |
+| 1.4458        | 5.0   | 35   | 2.5861          |
+| 1.3889        | 6.0   | 42   | 2.5646          |
+| 1.3527        | 7.0   | 49   | 2.5569          |
+| 1.2936        | 8.0   | 56   | 2.5532          |
+| 1.2855        | 9.0   | 63   | 2.5514          |
+| 1.2802        | 10.0  | 70   | 2.5520          |
+
+
+### Framework versions
+
+- PEFT 0.11.1
+- Transformers 4.40.1
+- Pytorch 2.2.0+cu121
+- Datasets 2.19.2
+- Tokenizers 0.19.1

adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "gate_proj",
+    "down_proj",
+    "k_proj",
+    "q_proj",
+    "o_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a38b465d1f590a1d840b09f8e51db8f7c1ad06eb5b0fb4214b92a84a7318eee
+size 19644912

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 5.48215128588288e+16,
+    "train_loss": 1.6741911002567835,
+    "train_runtime": 247.913,
+    "train_samples": 3959,
+    "train_samples_per_second": 17.062,
+    "train_steps_per_second": 0.282
+}

runs/Jun05_04-35-41_7bdd13775218/events.out.tfevents.1717562165.7bdd13775218.57453.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d859fc30deca884ddf93feb8be3a3201e8fc46b3a462aa1aac51e9efedd6872
+size 11560

special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:322a5f52ab5cab196761ab397a022d6fa3a2e1418585e532bb6efb2fedd2ae94
+size 17477501

tokenizer_config.json

@@ -0,0 +1,70 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "106": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{% if messages[0]['role'] == 'user' or messages[0]['role'] == 'system' %}{{ bos_token }}{% endif %}{% for message in messages %}{{ '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n' }}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% elif messages[-1]['role'] == 'assistant' %}{{ eos_token }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "legacy": null,
+  "model_max_length": 2048,
+  "pad_token": "<pad>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 5.48215128588288e+16,
+    "train_loss": 1.6741911002567835,
+    "train_runtime": 247.913,
+    "train_samples": 3959,
+    "train_samples_per_second": 17.062,
+    "train_steps_per_second": 0.282
+}

trainer_state.json

@@ -0,0 +1,215 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 70,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 2.171875,
+      "learning_rate": 2.857142857142857e-05,
+      "loss": 3.0133,
+      "step": 1
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 1.96875,
+      "learning_rate": 0.00014285714285714287,
+      "loss": 2.9598,
+      "step": 5
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.0818610191345215,
+      "eval_runtime": 0.5063,
+      "eval_samples_per_second": 19.752,
+      "eval_steps_per_second": 1.975,
+      "step": 7
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 1.4765625,
+      "learning_rate": 0.00019888308262251285,
+      "loss": 2.584,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.779895067214966,
+      "eval_runtime": 0.53,
+      "eval_samples_per_second": 18.868,
+      "eval_steps_per_second": 1.887,
+      "step": 14
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 1.2578125,
+      "learning_rate": 0.00019214762118704076,
+      "loss": 2.1727,
+      "step": 15
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.9609375,
+      "learning_rate": 0.00017971325072229226,
+      "loss": 1.8743,
+      "step": 20
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.6390492916107178,
+      "eval_runtime": 0.5412,
+      "eval_samples_per_second": 18.478,
+      "eval_steps_per_second": 1.848,
+      "step": 21
+    },
+    {
+      "epoch": 3.571428571428571,
+      "grad_norm": 0.609375,
+      "learning_rate": 0.00016234898018587337,
+      "loss": 1.6598,
+      "step": 25
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.6230721473693848,
+      "eval_runtime": 0.5473,
+      "eval_samples_per_second": 18.272,
+      "eval_steps_per_second": 1.827,
+      "step": 28
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 0.455078125,
+      "learning_rate": 0.00014112871031306119,
+      "loss": 1.5423,
+      "step": 30
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.5703125,
+      "learning_rate": 0.00011736481776669306,
+      "loss": 1.4458,
+      "step": 35
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.5861191749572754,
+      "eval_runtime": 0.5498,
+      "eval_samples_per_second": 18.188,
+      "eval_steps_per_second": 1.819,
+      "step": 35
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 0.404296875,
+      "learning_rate": 9.252699064135758e-05,
+      "loss": 1.3889,
+      "step": 40
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.5646462440490723,
+      "eval_runtime": 0.5474,
+      "eval_samples_per_second": 18.269,
+      "eval_steps_per_second": 1.827,
+      "step": 42
+    },
+    {
+      "epoch": 6.428571428571429,
+      "grad_norm": 0.34765625,
+      "learning_rate": 6.815133497483157e-05,
+      "loss": 1.3527,
+      "step": 45
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.556933879852295,
+      "eval_runtime": 0.5441,
+      "eval_samples_per_second": 18.378,
+      "eval_steps_per_second": 1.838,
+      "step": 49
+    },
+    {
+      "epoch": 7.142857142857143,
+      "grad_norm": 0.345703125,
+      "learning_rate": 4.574537361342407e-05,
+      "loss": 1.3118,
+      "step": 50
+    },
+    {
+      "epoch": 7.857142857142857,
+      "grad_norm": 0.44921875,
+      "learning_rate": 2.669481281701739e-05,
+      "loss": 1.2936,
+      "step": 55
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.5531744956970215,
+      "eval_runtime": 0.5391,
+      "eval_samples_per_second": 18.55,
+      "eval_steps_per_second": 1.855,
+      "step": 56
+    },
+    {
+      "epoch": 8.571428571428571,
+      "grad_norm": 0.32421875,
+      "learning_rate": 1.2177842662977135e-05,
+      "loss": 1.2855,
+      "step": 60
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.551410436630249,
+      "eval_runtime": 0.5405,
+      "eval_samples_per_second": 18.501,
+      "eval_steps_per_second": 1.85,
+      "step": 63
+    },
+    {
+      "epoch": 9.285714285714286,
+      "grad_norm": 0.244140625,
+      "learning_rate": 3.092271377092215e-06,
+      "loss": 1.2767,
+      "step": 65
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.2431640625,
+      "learning_rate": 0.0,
+      "loss": 1.2802,
+      "step": 70
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.5520482063293457,
+      "eval_runtime": 0.5385,
+      "eval_samples_per_second": 18.569,
+      "eval_steps_per_second": 1.857,
+      "step": 70
+    },
+    {
+      "epoch": 10.0,
+      "step": 70,
+      "total_flos": 5.48215128588288e+16,
+      "train_loss": 1.6741911002567835,
+      "train_runtime": 247.913,
+      "train_samples_per_second": 17.062,
+      "train_steps_per_second": 0.282
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 70,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 100,
+  "total_flos": 5.48215128588288e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11d91c8af61fbc32d8bd78e66a477a14a1bd6371d78bee49c309c30c00443ff5
+size 5176