Model save

README.md

@@ -0,0 +1,71 @@
+---
+license: mit
+base_model: HuggingFaceH4/zephyr-7b-beta
+tags:
+- generated_from_trainer
+model-index:
+- name: zephyr-7b-ikat
+  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. -->
+
+# zephyr-7b-ikat
+
+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: 0.5166
+
+## 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: 8
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 256
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.7437        | 0.62  | 17   | 0.6867          |
+| 0.6372        | 1.63  | 35   | 0.6215          |
+| 0.6078        | 2.64  | 53   | 0.5859          |
+| 0.5724        | 3.62  | 70   | 0.5625          |
+| 0.5613        | 4.63  | 88   | 0.5448          |
+| 0.5427        | 5.64  | 106  | 0.5337          |
+| 0.5388        | 6.62  | 123  | 0.5274          |
+| 0.5284        | 7.63  | 141  | 0.5229          |
+| 0.5285        | 8.64  | 159  | 0.5188          |
+| 0.5222        | 9.61  | 176  | 0.5165          |
+
+
+### Framework versions
+
+- Transformers 4.35.0
+- Pytorch 2.1.1+cu118
+- Datasets 2.14.6
+- Tokenizers 0.14.1

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": 16,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "v_proj",
+    "o_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f166319ea108369065349f0caf68c4d2e72c1d26dfe9e6b5c242a09e25ddb31d
+size 109086672

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 9.61,
+    "eval_loss": 0.5165773034095764,
+    "eval_runtime": 30.0245,
+    "eval_samples": 200,
+    "eval_samples_per_second": 6.661,
+    "eval_steps_per_second": 0.833,
+    "train_loss": 0.5895595401525497,
+    "train_runtime": 27383.9371,
+    "train_samples": 7016,
+    "train_samples_per_second": 2.562,
+    "train_steps_per_second": 0.01
+}

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "HuggingFaceH4/zephyr-7b-beta",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pad_token_id": 2,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.35.0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.61,
+    "eval_loss": 0.5165773034095764,
+    "eval_runtime": 30.0245,
+    "eval_samples": 200,
+    "eval_samples_per_second": 6.661,
+    "eval_steps_per_second": 0.833
+}

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_config.json

@@ -0,0 +1,46 @@
+{
+  "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": 2048,
+  "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": 9.61,
+    "train_loss": 0.5895595401525497,
+    "train_runtime": 27383.9371,
+    "train_samples": 7016,
+    "train_samples_per_second": 2.562,
+    "train_steps_per_second": 0.01
+}

trainer_state.json

@@ -0,0 +1,324 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 9.612756264236902,
+  "eval_steps": 500,
+  "global_step": 176,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9999323080037623e-05,
+      "loss": 0.8679,
+      "step": 1
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9983081582712684e-05,
+      "loss": 0.8526,
+      "step": 5
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9932383577419432e-05,
+      "loss": 0.8014,
+      "step": 10
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9848077530122083e-05,
+      "loss": 0.7437,
+      "step": 15
+    },
+    {
+      "epoch": 0.62,
+      "eval_loss": 0.6867304444313049,
+      "eval_runtime": 30.7118,
+      "eval_samples_per_second": 6.512,
+      "eval_steps_per_second": 0.814,
+      "step": 17
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.973044870579824e-05,
+      "loss": 0.6989,
+      "step": 20
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.957989512315489e-05,
+      "loss": 0.6673,
+      "step": 25
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.9396926207859085e-05,
+      "loss": 0.6539,
+      "step": 30
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.9182161068802742e-05,
+      "loss": 0.6372,
+      "step": 35
+    },
+    {
+      "epoch": 1.63,
+      "eval_loss": 0.621515691280365,
+      "eval_runtime": 30.0651,
+      "eval_samples_per_second": 6.652,
+      "eval_steps_per_second": 0.832,
+      "step": 35
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 1.8936326403234125e-05,
+      "loss": 0.6243,
+      "step": 40
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.866025403784439e-05,
+      "loss": 0.6186,
+      "step": 45
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 1.8354878114129368e-05,
+      "loss": 0.6078,
+      "step": 50
+    },
+    {
+      "epoch": 2.64,
+      "eval_loss": 0.5859256386756897,
+      "eval_runtime": 30.0404,
+      "eval_samples_per_second": 6.658,
+      "eval_steps_per_second": 0.832,
+      "step": 53
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 1.802123192755044e-05,
+      "loss": 0.5942,
+      "step": 55
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 1.766044443118978e-05,
+      "loss": 0.5896,
+      "step": 60
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 1.7273736415730488e-05,
+      "loss": 0.5848,
+      "step": 65
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 1.686241637868734e-05,
+      "loss": 0.5724,
+      "step": 70
+    },
+    {
+      "epoch": 3.62,
+      "eval_loss": 0.562507688999176,
+      "eval_runtime": 30.0633,
+      "eval_samples_per_second": 6.653,
+      "eval_steps_per_second": 0.832,
+      "step": 70
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 1.6427876096865394e-05,
+      "loss": 0.5703,
+      "step": 75
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 1.5971585917027864e-05,
+      "loss": 0.5612,
+      "step": 80
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 1.5495089780708062e-05,
+      "loss": 0.5613,
+      "step": 85
+    },
+    {
+      "epoch": 4.63,
+      "eval_loss": 0.544775128364563,
+      "eval_runtime": 30.0298,
+      "eval_samples_per_second": 6.66,
+      "eval_steps_per_second": 0.833,
+      "step": 88
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.5509,
+      "step": 90
+    },
+    {
+      "epoch": 5.24,
+      "learning_rate": 1.4487991802004625e-05,
+      "loss": 0.5486,
+      "step": 95
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 1.396079766039157e-05,
+      "loss": 0.5485,
+      "step": 100
+    },
+    {
+      "epoch": 5.6,
+      "learning_rate": 1.342020143325669e-05,
+      "loss": 0.5427,
+      "step": 105
+    },
+    {
+      "epoch": 5.64,
+      "eval_loss": 0.5336735844612122,
+      "eval_runtime": 30.037,
+      "eval_samples_per_second": 6.658,
+      "eval_steps_per_second": 0.832,
+      "step": 106
+    },
+    {
+      "epoch": 6.14,
+      "learning_rate": 1.2868032327110904e-05,
+      "loss": 0.5399,
+      "step": 110
+    },
+    {
+      "epoch": 6.32,
+      "learning_rate": 1.2306158707424402e-05,
+      "loss": 0.5376,
+      "step": 115
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 1.1736481776669307e-05,
+      "loss": 0.5388,
+      "step": 120
+    },
+    {
+      "epoch": 6.62,
+      "eval_loss": 0.5273823142051697,
+      "eval_runtime": 30.0618,
+      "eval_samples_per_second": 6.653,
+      "eval_steps_per_second": 0.832,
+      "step": 123
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 1.1160929141252303e-05,
+      "loss": 0.5321,
+      "step": 125
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 1.0581448289104759e-05,
+      "loss": 0.5323,
+      "step": 130
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 1e-05,
+      "loss": 0.5326,
+      "step": 135
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 9.418551710895243e-06,
+      "loss": 0.5284,
+      "step": 140
+    },
+    {
+      "epoch": 7.63,
+      "eval_loss": 0.5229138731956482,
+      "eval_runtime": 30.0491,
+      "eval_samples_per_second": 6.656,
+      "eval_steps_per_second": 0.832,
+      "step": 141
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 8.839070858747697e-06,
+      "loss": 0.529,
+      "step": 145
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 8.263518223330698e-06,
+      "loss": 0.5277,
+      "step": 150
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 7.6938412925756e-06,
+      "loss": 0.5285,
+      "step": 155
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 0.5188391208648682,
+      "eval_runtime": 30.0316,
+      "eval_samples_per_second": 6.66,
+      "eval_steps_per_second": 0.832,
+      "step": 159
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 7.131967672889101e-06,
+      "loss": 0.5219,
+      "step": 160
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 6.579798566743314e-06,
+      "loss": 0.5249,
+      "step": 165
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 6.039202339608432e-06,
+      "loss": 0.5249,
+      "step": 170
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 5.512008197995379e-06,
+      "loss": 0.5222,
+      "step": 175
+    },
+    {
+      "epoch": 9.61,
+      "eval_loss": 0.5165340304374695,
+      "eval_runtime": 30.0644,
+      "eval_samples_per_second": 6.652,
+      "eval_steps_per_second": 0.832,
+      "step": 176
+    },
+    {
+      "epoch": 9.61,
+      "step": 176,
+      "total_flos": 1.59704785354752e+16,
+      "train_loss": 0.5895595401525497,
+      "train_runtime": 27383.9371,
+      "train_samples_per_second": 2.562,
+      "train_steps_per_second": 0.01
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 270,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1.59704785354752e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7afc8cf8129522d9383886b34de8fc50050e9d86f2a554c68ff58e4b0be77d7
+size 5624