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 +1,82 @@
+---
+license: gemma
+library_name: peft
+tags:
+- trl
+- sft
+- generated_from_trainer
+base_model: google/gemma-7b
+datasets:
+- generator
+model-index:
+- name: gemma-7b-sft-qlora-no-robots15
+  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. -->
+
+# gemma-7b-sft-qlora-no-robots15
+
+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: 1.2830
+
+## 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: 2
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 15
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 21.906        | 0.91  | 5    | 7.6533          |
+| 13.5603       | 2.0   | 11   | 6.6442          |
+| 10.2605       | 2.91  | 16   | 6.0815          |
+| 9.9129        | 4.0   | 22   | 3.1148          |
+| 4.5895        | 4.91  | 27   | 1.6583          |
+| 1.6316        | 6.0   | 33   | 1.4155          |
+| 1.4115        | 6.91  | 38   | 1.3543          |
+| 1.2971        | 8.0   | 44   | 1.3133          |
+| 1.1321        | 8.91  | 49   | 1.2903          |
+| 0.9739        | 10.0  | 55   | 1.2820          |
+| 0.917         | 10.91 | 60   | 1.2888          |
+| 0.8541        | 12.0  | 66   | 1.2781          |
+| 0.8659        | 12.91 | 71   | 1.2892          |
+| 0.8354        | 13.64 | 75   | 1.2830          |
+
+
+### Framework versions
+
+- PEFT 0.7.1
+- Transformers 4.39.3
+- Pytorch 2.2.2+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

adapter_config.json

@@ -0,0 +1,31 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-7b",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "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": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "o_proj",
+    "k_proj",
+    "up_proj",
+    "gate_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be6e49bde386e61c34c9dd54410a8a0f5f837bcefb36a8042ba70849c4ef6788
+size 100060536

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 13.64,
+    "train_loss": 4.788167775472005,
+    "train_runtime": 283.7979,
+    "train_samples": 926,
+    "train_samples_per_second": 4.651,
+    "train_steps_per_second": 0.264
+}

runs/Apr11_11-49-12_deep-diver-main-gentle-ape-1-0-0/events.out.tfevents.1712850662.deep-diver-main-gentle-ape-1-0-0.519.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1414e16a432fdf84a812adb4b7211a7476591f30e18895bbf899013f000f5d0a
+size 12771

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 sha256:22449cb9ef4bad0db7dd93b46ddff7ab7d6a654dd4f903e130ddb6361eac3af5
+size 17477473

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 +1,8 @@
+{
+    "epoch": 13.64,
+    "train_loss": 4.788167775472005,
+    "train_runtime": 283.7979,
+    "train_samples": 926,
+    "train_samples_per_second": 4.651,
+    "train_steps_per_second": 0.264
+}

trainer_state.json

@@ -0,0 +1,254 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 13.636363636363637,
+  "eval_steps": 500,
+  "global_step": 75,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.18,
+      "grad_norm": 220.0,
+      "learning_rate": 2.5e-05,
+      "loss": 24.9691,
+      "step": 1
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 35.25,
+      "learning_rate": 0.000125,
+      "loss": 21.906,
+      "step": 5
+    },
+    {
+      "epoch": 0.91,
+      "eval_loss": 7.653346538543701,
+      "eval_runtime": 0.5608,
+      "eval_samples_per_second": 3.567,
+      "eval_steps_per_second": 1.783,
+      "step": 5
+    },
+    {
+      "epoch": 1.82,
+      "grad_norm": 9.3125,
+      "learning_rate": 0.00019956059820218982,
+      "loss": 13.5603,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 6.644189357757568,
+      "eval_runtime": 0.5604,
+      "eval_samples_per_second": 3.569,
+      "eval_steps_per_second": 1.784,
+      "step": 11
+    },
+    {
+      "epoch": 2.73,
+      "grad_norm": 4.625,
+      "learning_rate": 0.00019466156752904343,
+      "loss": 10.2605,
+      "step": 15
+    },
+    {
+      "epoch": 2.91,
+      "eval_loss": 6.081549644470215,
+      "eval_runtime": 0.5806,
+      "eval_samples_per_second": 3.445,
+      "eval_steps_per_second": 1.722,
+      "step": 16
+    },
+    {
+      "epoch": 3.64,
+      "grad_norm": 15.125,
+      "learning_rate": 0.00018458320592590975,
+      "loss": 9.9129,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.114753484725952,
+      "eval_runtime": 0.5656,
+      "eval_samples_per_second": 3.536,
+      "eval_steps_per_second": 1.768,
+      "step": 22
+    },
+    {
+      "epoch": 4.55,
+      "grad_norm": 2.96875,
+      "learning_rate": 0.00016987694277788417,
+      "loss": 4.5895,
+      "step": 25
+    },
+    {
+      "epoch": 4.91,
+      "eval_loss": 1.6582958698272705,
+      "eval_runtime": 0.5842,
+      "eval_samples_per_second": 3.424,
+      "eval_steps_per_second": 1.712,
+      "step": 27
+    },
+    {
+      "epoch": 5.45,
+      "grad_norm": 3.171875,
+      "learning_rate": 0.0001513474193514842,
+      "loss": 1.6316,
+      "step": 30
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.415539264678955,
+      "eval_runtime": 0.5662,
+      "eval_samples_per_second": 3.532,
+      "eval_steps_per_second": 1.766,
+      "step": 33
+    },
+    {
+      "epoch": 6.36,
+      "grad_norm": 7.1875,
+      "learning_rate": 0.0001300084635000341,
+      "loss": 1.4115,
+      "step": 35
+    },
+    {
+      "epoch": 6.91,
+      "eval_loss": 1.3542958498001099,
+      "eval_runtime": 0.58,
+      "eval_samples_per_second": 3.448,
+      "eval_steps_per_second": 1.724,
+      "step": 38
+    },
+    {
+      "epoch": 7.27,
+      "grad_norm": 1.296875,
+      "learning_rate": 0.0001070276188945293,
+      "loss": 1.2971,
+      "step": 40
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 1.313336968421936,
+      "eval_runtime": 0.5751,
+      "eval_samples_per_second": 3.478,
+      "eval_steps_per_second": 1.739,
+      "step": 44
+    },
+    {
+      "epoch": 8.18,
+      "grad_norm": 1.6171875,
+      "learning_rate": 8.366226381814697e-05,
+      "loss": 1.1321,
+      "step": 45
+    },
+    {
+      "epoch": 8.91,
+      "eval_loss": 1.2903474569320679,
+      "eval_runtime": 0.5848,
+      "eval_samples_per_second": 3.42,
+      "eval_steps_per_second": 1.71,
+      "step": 49
+    },
+    {
+      "epoch": 9.09,
+      "grad_norm": 0.9375,
+      "learning_rate": 6.119081473277501e-05,
+      "loss": 1.062,
+      "step": 50
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.90234375,
+      "learning_rate": 4.084277875864776e-05,
+      "loss": 0.9739,
+      "step": 55
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 1.2820332050323486,
+      "eval_runtime": 0.5646,
+      "eval_samples_per_second": 3.542,
+      "eval_steps_per_second": 1.771,
+      "step": 55
+    },
+    {
+      "epoch": 10.91,
+      "grad_norm": 1.2109375,
+      "learning_rate": 2.3731482188961818e-05,
+      "loss": 0.917,
+      "step": 60
+    },
+    {
+      "epoch": 10.91,
+      "eval_loss": 1.2888375520706177,
+      "eval_runtime": 0.5836,
+      "eval_samples_per_second": 3.427,
+      "eval_steps_per_second": 1.714,
+      "step": 60
+    },
+    {
+      "epoch": 11.82,
+      "grad_norm": 0.609375,
+      "learning_rate": 1.0793155744261351e-05,
+      "loss": 0.8541,
+      "step": 65
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 1.27806556224823,
+      "eval_runtime": 0.5683,
+      "eval_samples_per_second": 3.519,
+      "eval_steps_per_second": 1.76,
+      "step": 66
+    },
+    {
+      "epoch": 12.73,
+      "grad_norm": 0.6796875,
+      "learning_rate": 2.735709467518699e-06,
+      "loss": 0.8659,
+      "step": 70
+    },
+    {
+      "epoch": 12.91,
+      "eval_loss": 1.2891546487808228,
+      "eval_runtime": 0.7414,
+      "eval_samples_per_second": 2.698,
+      "eval_steps_per_second": 1.349,
+      "step": 71
+    },
+    {
+      "epoch": 13.64,
+      "grad_norm": 0.515625,
+      "learning_rate": 0.0,
+      "loss": 0.8354,
+      "step": 75
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 1.2830009460449219,
+      "eval_runtime": 0.5659,
+      "eval_samples_per_second": 3.534,
+      "eval_steps_per_second": 1.767,
+      "step": 75
+    },
+    {
+      "epoch": 13.64,
+      "step": 75,
+      "total_flos": 1.1503415671442637e+17,
+      "train_loss": 4.788167775472005,
+      "train_runtime": 283.7979,
+      "train_samples_per_second": 4.651,
+      "train_steps_per_second": 0.264
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 75,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 15,
+  "save_steps": 100,
+  "total_flos": 1.1503415671442637e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffbfc1f7707f8c234a55a9fc0b314774d4ec0938f941afe549a7d87e7a4373e1
+size 5112