upload model checkpoints

.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

@@ -1,3 +1,76 @@
----
-license: gemma
----
+---
+library_name: peft
+tags:
+- alignment-handbook
+- generated_from_trainer
+datasets:
+- llama-duo/synth_summarize_dataset_dedup
+base_model: google/gemma-7b
+model-index:
+- name: gemma7b-summarize-gemini1_5flash-8k
+  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. -->
+
+# gemma7b-summarize-gemini1_5flash-8k
+
+This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the llama-duo/synth_summarize_dataset_dedup dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.8396
+
+## 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: 4
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- 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: 10
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 31.311        | 0.9630 | 13   | 11.0230         |
+| 19.0651       | 2.0    | 27   | 7.4342          |
+| 11.34         | 2.9630 | 40   | 6.8118          |
+| 3.0136        | 4.0    | 54   | 3.6308          |
+| 1.7786        | 4.9630 | 67   | 3.0973          |
+| 1.4865        | 6.0    | 81   | 2.9241          |
+| 1.4036        | 6.9630 | 94   | 2.8645          |
+| 1.3424        | 8.0    | 108  | 2.8510          |
+| 1.3298        | 8.9630 | 121  | 2.8410          |
+| 1.3245        | 9.6296 | 130  | 2.8396          |
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.0
+- Pytorch 2.1.2+cu121
+- Datasets 2.18.0
+- Tokenizers 0.19.1

adapter_config.json

@@ -0,0 +1,29 @@
+{
+  "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,
+  "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": [
+    "v_proj",
+    "q_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 sha256:6229fc9ac1efc88d558f0bd2f363a10205884dcaf1e97306d89c2c0e99209d92
+size 6437384

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 9.62962962962963,
+    "eval_loss": 2.839648723602295,
+    "eval_runtime": 0.2362,
+    "eval_samples": 25,
+    "eval_samples_per_second": 42.345,
+    "eval_steps_per_second": 4.234,
+    "total_flos": 3.9639312421093376e+17,
+    "train_loss": 8.499957752227782,
+    "train_runtime": 318.2463,
+    "train_samples": 7919,
+    "train_samples_per_second": 26.238,
+    "train_steps_per_second": 0.408
+}

config.json

@@ -0,0 +1,44 @@
+{
+  "_name_or_path": "google/gemma-7b",
+  "architectures": [
+    "GemmaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 2,
+  "eos_token_id": 1,
+  "head_dim": 256,
+  "hidden_act": "gelu",
+  "hidden_activation": null,
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 24576,
+  "max_position_embeddings": 8192,
+  "model_type": "gemma",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 16,
+  "pad_token_id": 0,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "nf4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.0",
+  "use_cache": true,
+  "vocab_size": 256000
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.62962962962963,
+    "eval_loss": 2.839648723602295,
+    "eval_runtime": 0.2362,
+    "eval_samples": 25,
+    "eval_samples_per_second": 42.345,
+    "eval_steps_per_second": 4.234
+}

runs/Jun13_16-13-44_gpu1-1/events.out.tfevents.1718266897.gpu1-1.616100.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3812375f6d1b4bd411141a2061ec803a297339d2e6bdd8be0f29b0208838da75
+size 14095

runs/Jun13_16-13-44_gpu1-1/events.out.tfevents.1718267216.gpu1-1.616100.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a8bbc3525c5d122ee5fa7b35cccc4556ae19f06a3099ff3aad8c16b426ea77a
+size 359

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": 9.62962962962963,
+    "total_flos": 3.9639312421093376e+17,
+    "train_loss": 8.499957752227782,
+    "train_runtime": 318.2463,
+    "train_samples": 7919,
+    "train_samples_per_second": 26.238,
+    "train_steps_per_second": 0.408
+}

trainer_state.json

@@ -0,0 +1,299 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 9.62962962962963,
+  "eval_steps": 500,
+  "global_step": 130,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07407407407407407,
+      "grad_norm": 178.0,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 48.7098,
+      "step": 1
+    },
+    {
+      "epoch": 0.37037037037037035,
+      "grad_norm": 141.0,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": 49.2488,
+      "step": 5
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 17.75,
+      "learning_rate": 0.00015384615384615385,
+      "loss": 31.311,
+      "step": 10
+    },
+    {
+      "epoch": 0.9629629629629629,
+      "eval_loss": 11.023000717163086,
+      "eval_runtime": 0.2736,
+      "eval_samples_per_second": 36.555,
+      "eval_steps_per_second": 3.655,
+      "step": 13
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 10.0,
+      "learning_rate": 0.00019985583705641418,
+      "loss": 23.1246,
+      "step": 15
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 4.46875,
+      "learning_rate": 0.00019823877374156647,
+      "loss": 20.1941,
+      "step": 20
+    },
+    {
+      "epoch": 1.8518518518518519,
+      "grad_norm": 6.0625,
+      "learning_rate": 0.00019485364419471454,
+      "loss": 19.0651,
+      "step": 25
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 7.434217929840088,
+      "eval_runtime": 0.2352,
+      "eval_samples_per_second": 42.524,
+      "eval_steps_per_second": 4.252,
+      "step": 27
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 11.5,
+      "learning_rate": 0.0001897613727639014,
+      "loss": 17.6692,
+      "step": 30
+    },
+    {
+      "epoch": 2.5925925925925926,
+      "grad_norm": 18.625,
+      "learning_rate": 0.00018305360832480117,
+      "loss": 15.2716,
+      "step": 35
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 25.75,
+      "learning_rate": 0.00017485107481711012,
+      "loss": 11.34,
+      "step": 40
+    },
+    {
+      "epoch": 2.962962962962963,
+      "eval_loss": 6.81179666519165,
+      "eval_runtime": 0.2444,
+      "eval_samples_per_second": 40.915,
+      "eval_steps_per_second": 4.091,
+      "step": 40
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 26.125,
+      "learning_rate": 0.0001653013984983585,
+      "loss": 6.3302,
+      "step": 45
+    },
+    {
+      "epoch": 3.7037037037037037,
+      "grad_norm": 7.375,
+      "learning_rate": 0.00015457645101945046,
+      "loss": 3.0136,
+      "step": 50
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.630751848220825,
+      "eval_runtime": 0.2376,
+      "eval_samples_per_second": 42.085,
+      "eval_steps_per_second": 4.208,
+      "step": 54
+    },
+    {
+      "epoch": 4.074074074074074,
+      "grad_norm": 5.59375,
+      "learning_rate": 0.00014286925614030542,
+      "loss": 2.4006,
+      "step": 55
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 2.859375,
+      "learning_rate": 0.0001303905157574247,
+      "loss": 2.0441,
+      "step": 60
+    },
+    {
+      "epoch": 4.814814814814815,
+      "grad_norm": 1.046875,
+      "learning_rate": 0.00011736481776669306,
+      "loss": 1.7786,
+      "step": 65
+    },
+    {
+      "epoch": 4.962962962962963,
+      "eval_loss": 3.097276210784912,
+      "eval_runtime": 0.2571,
+      "eval_samples_per_second": 38.893,
+      "eval_steps_per_second": 3.889,
+      "step": 67
+    },
+    {
+      "epoch": 5.185185185185185,
+      "grad_norm": 0.703125,
+      "learning_rate": 0.00010402659401094152,
+      "loss": 1.6515,
+      "step": 70
+    },
+    {
+      "epoch": 5.555555555555555,
+      "grad_norm": 0.69921875,
+      "learning_rate": 9.061590105968208e-05,
+      "loss": 1.5441,
+      "step": 75
+    },
+    {
+      "epoch": 5.925925925925926,
+      "grad_norm": 1.5,
+      "learning_rate": 7.73740997570278e-05,
+      "loss": 1.4865,
+      "step": 80
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.9240853786468506,
+      "eval_runtime": 0.2376,
+      "eval_samples_per_second": 42.082,
+      "eval_steps_per_second": 4.208,
+      "step": 81
+    },
+    {
+      "epoch": 6.296296296296296,
+      "grad_norm": 1.2421875,
+      "learning_rate": 6.453951129574644e-05,
+      "loss": 1.448,
+      "step": 85
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 0.5,
+      "learning_rate": 5.234312799786921e-05,
+      "loss": 1.4036,
+      "step": 90
+    },
+    {
+      "epoch": 6.962962962962963,
+      "eval_loss": 2.864516019821167,
+      "eval_runtime": 0.263,
+      "eval_samples_per_second": 38.017,
+      "eval_steps_per_second": 3.802,
+      "step": 94
+    },
+    {
+      "epoch": 7.037037037037037,
+      "grad_norm": 0.484375,
+      "learning_rate": 4.100445599768774e-05,
+      "loss": 1.3829,
+      "step": 95
+    },
+    {
+      "epoch": 7.407407407407407,
+      "grad_norm": 0.4765625,
+      "learning_rate": 3.072756464904006e-05,
+      "loss": 1.3704,
+      "step": 100
+    },
+    {
+      "epoch": 7.777777777777778,
+      "grad_norm": 0.51953125,
+      "learning_rate": 2.1697413758237784e-05,
+      "loss": 1.3424,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.851011276245117,
+      "eval_runtime": 0.2392,
+      "eval_samples_per_second": 41.815,
+      "eval_steps_per_second": 4.181,
+      "step": 108
+    },
+    {
+      "epoch": 8.148148148148149,
+      "grad_norm": 0.447265625,
+      "learning_rate": 1.4076524743778319e-05,
+      "loss": 1.3438,
+      "step": 110
+    },
+    {
+      "epoch": 8.518518518518519,
+      "grad_norm": 0.451171875,
+      "learning_rate": 8.002055634117578e-06,
+      "loss": 1.3448,
+      "step": 115
+    },
+    {
+      "epoch": 8.88888888888889,
+      "grad_norm": 0.54296875,
+      "learning_rate": 3.5833325466437694e-06,
+      "loss": 1.3298,
+      "step": 120
+    },
+    {
+      "epoch": 8.962962962962964,
+      "eval_loss": 2.8409743309020996,
+      "eval_runtime": 0.2622,
+      "eval_samples_per_second": 38.132,
+      "eval_steps_per_second": 3.813,
+      "step": 121
+    },
+    {
+      "epoch": 9.25925925925926,
+      "grad_norm": 0.451171875,
+      "learning_rate": 8.998820754091531e-07,
+      "loss": 1.3428,
+      "step": 125
+    },
+    {
+      "epoch": 9.62962962962963,
+      "grad_norm": 0.40625,
+      "learning_rate": 0.0,
+      "loss": 1.3245,
+      "step": 130
+    },
+    {
+      "epoch": 9.62962962962963,
+      "eval_loss": 2.839648723602295,
+      "eval_runtime": 0.2376,
+      "eval_samples_per_second": 42.081,
+      "eval_steps_per_second": 4.208,
+      "step": 130
+    },
+    {
+      "epoch": 9.62962962962963,
+      "step": 130,
+      "total_flos": 3.9639312421093376e+17,
+      "train_loss": 8.499957752227782,
+      "train_runtime": 318.2463,
+      "train_samples_per_second": 26.238,
+      "train_steps_per_second": 0.408
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 130,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 100,
+  "total_flos": 3.9639312421093376e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d5dbae39d16366f390d08c76e743cee69ed721f47930e89c14ef648062bc188
+size 5176