Model save

README.md

@@ -0,0 +1,65 @@
+---
+license: mit
+base_model: HuggingFaceH4/zephyr-7b-beta
+tags:
+- generated_from_trainer
+model-index:
+- name: sft-zephyr-7b-beta-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. -->
+
+# sft-zephyr-7b-beta-v1
+
+This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4405
+
+## 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: 2e-05
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.05
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.8404        | 0.19  | 50   | 0.7357          |
+| 0.5565        | 0.37  | 100  | 0.5301          |
+| 0.5284        | 1.14  | 150  | 0.4864          |
+| 0.4912        | 1.33  | 200  | 0.4633          |
+| 0.4473        | 2.1   | 250  | 0.4545          |
+| 0.4285        | 2.29  | 300  | 0.4483          |
+
+
+### Framework versions
+
+- Transformers 4.35.2
+- Pytorch 2.1.0
+- Datasets 2.15.0
+- Tokenizers 0.15.0

adapter_config.json

@@ -0,0 +1,25 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "HuggingFaceH4/zephyr-7b-beta",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4043efd316e6c8591ddf522aab3caddae333fdb275ba9a1cbd2b45adfff8a345
+size 54560368

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 2.41,
+    "eval_loss": 0.4404671788215637,
+    "eval_runtime": 10.292,
+    "eval_samples": 120,
+    "eval_samples_per_second": 11.66,
+    "eval_steps_per_second": 2.915,
+    "train_loss": 0.5767455007936861,
+    "train_runtime": 1186.2419,
+    "train_samples": 1076,
+    "train_samples_per_second": 2.721,
+    "train_steps_per_second": 0.68
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.41,
+    "eval_loss": 0.4404671788215637,
+    "eval_runtime": 10.292,
+    "eval_samples": 120,
+    "eval_samples_per_second": 11.66,
+    "eval_steps_per_second": 2.915
+}

runs/Nov17_20-55-39_7a59b30c842e/events.out.tfevents.1700254679.7a59b30c842e.13548.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58e6cfe46e8fc46e63aa43763d938afc57d38010fac08253c4b73a33b87cc775
+size 11570

runs/Nov17_20-55-39_7a59b30c842e/events.out.tfevents.1700255876.7a59b30c842e.13548.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0a79d9defde9adbc8f954b5d565a0a33cc5c5324e4d1e744d3c90655cb248fd
+size 359

special_tokens_map.json

@@ -0,0 +1,35 @@
+{
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,48 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>"
+  ],
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "truncation_side": "left",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.41,
+    "train_loss": 0.5767455007936861,
+    "train_runtime": 1186.2419,
+    "train_samples": 1076,
+    "train_samples_per_second": 2.721,
+    "train_steps_per_second": 0.68
+}

trainer_state.json

@@ -0,0 +1,280 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.412639405204461,
+  "eval_steps": 50,
+  "global_step": 333,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.878048780487805e-07,
+      "loss": 1.1781,
+      "step": 1
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.8780487804878055e-06,
+      "loss": 1.1487,
+      "step": 10
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.756097560975611e-06,
+      "loss": 1.1115,
+      "step": 20
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.4634146341463415e-05,
+      "loss": 1.0121,
+      "step": 30
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9512195121951222e-05,
+      "loss": 0.9286,
+      "step": 40
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9993188419095562e-05,
+      "loss": 0.8404,
+      "step": 50
+    },
+    {
+      "epoch": 0.19,
+      "eval_loss": 0.7357456088066101,
+      "eval_runtime": 10.273,
+      "eval_samples_per_second": 11.681,
+      "eval_steps_per_second": 2.92,
+      "step": 50
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9969654126764183e-05,
+      "loss": 0.7129,
+      "step": 60
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.992935260059287e-05,
+      "loss": 0.6721,
+      "step": 70
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9872351620777883e-05,
+      "loss": 0.6196,
+      "step": 80
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9798747053108098e-05,
+      "loss": 0.5785,
+      "step": 90
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9708662687735316e-05,
+      "loss": 0.5565,
+      "step": 100
+    },
+    {
+      "epoch": 0.37,
+      "eval_loss": 0.5301057696342468,
+      "eval_runtime": 10.2864,
+      "eval_samples_per_second": 11.666,
+      "eval_steps_per_second": 2.916,
+      "step": 100
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9602250030980657e-05,
+      "loss": 0.5421,
+      "step": 110
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.947968805052712e-05,
+      "loss": 0.5197,
+      "step": 120
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.934118287442689e-05,
+      "loss": 0.508,
+      "step": 130
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.9186967444429613e-05,
+      "loss": 0.5097,
+      "step": 140
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.901730112421468e-05,
+      "loss": 0.5284,
+      "step": 150
+    },
+    {
+      "epoch": 1.14,
+      "eval_loss": 0.4864448308944702,
+      "eval_runtime": 10.291,
+      "eval_samples_per_second": 11.661,
+      "eval_steps_per_second": 2.915,
+      "step": 150
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.8832469263186352e-05,
+      "loss": 0.5095,
+      "step": 160
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.8632782716565438e-05,
+      "loss": 0.4823,
+      "step": 170
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.841857732258457e-05,
+      "loss": 0.4792,
+      "step": 180
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.8190213337666384e-05,
+      "loss": 0.4921,
+      "step": 190
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.7948074830534535e-05,
+      "loss": 0.4912,
+      "step": 200
+    },
+    {
+      "epoch": 1.33,
+      "eval_loss": 0.4633093774318695,
+      "eval_runtime": 10.2883,
+      "eval_samples_per_second": 11.664,
+      "eval_steps_per_second": 2.916,
+      "step": 200
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.7692569036276533e-05,
+      "loss": 0.4684,
+      "step": 210
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.742412567144476e-05,
+      "loss": 0.4928,
+      "step": 220
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.714319621134755e-05,
+      "loss": 0.4648,
+      "step": 230
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 1.685025313074582e-05,
+      "loss": 0.4673,
+      "step": 240
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 1.6545789109232247e-05,
+      "loss": 0.4473,
+      "step": 250
+    },
+    {
+      "epoch": 2.1,
+      "eval_loss": 0.45452865958213806,
+      "eval_runtime": 10.2907,
+      "eval_samples_per_second": 11.661,
+      "eval_steps_per_second": 2.915,
+      "step": 250
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 1.6230316202629393e-05,
+      "loss": 0.4551,
+      "step": 260
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 1.590436498180039e-05,
+      "loss": 0.4366,
+      "step": 270
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.556848364032052e-05,
+      "loss": 0.4399,
+      "step": 280
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 1.5223237072510433e-05,
+      "loss": 0.427,
+      "step": 290
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.4904987486286184e-05,
+      "loss": 0.4285,
+      "step": 300
+    },
+    {
+      "epoch": 2.29,
+      "eval_loss": 0.44827380776405334,
+      "eval_runtime": 10.2956,
+      "eval_samples_per_second": 11.655,
+      "eval_steps_per_second": 2.914,
+      "step": 300
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 1.454355889687053e-05,
+      "loss": 0.4233,
+      "step": 310
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 1.4174488827267032e-05,
+      "loss": 0.4339,
+      "step": 320
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 1.3798397989496549e-05,
+      "loss": 0.4381,
+      "step": 330
+    },
+    {
+      "epoch": 2.41,
+      "step": 333,
+      "total_flos": 5.790989170402918e+16,
+      "train_loss": 0.5767455007936861,
+      "train_runtime": 1186.2419,
+      "train_samples_per_second": 2.721,
+      "train_steps_per_second": 0.68
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 807,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 5.790989170402918e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75e1db26d33e0bbb23a35ece413c42f84f21154a34d690cc7ed6d255e921bcb6
+size 4600