Model save

README.md

@@ -0,0 +1,62 @@
+---
+license: mit
+base_model: HuggingFaceH4/mistral-7b-sft-beta
+tags:
+- generated_from_trainer
+model-index:
+- name: zephyr-7b-sft-lora
+  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-sft-lora
+
+This model is a fine-tuned version of [HuggingFaceH4/mistral-7b-sft-beta](https://huggingface.co/HuggingFaceH4/mistral-7b-sft-beta) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8835
+
+## 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: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 128
+- total_train_batch_size: 4096
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.8999        | 0.99  | 15   | 0.8835          |
+
+
+### Framework versions
+
+- Transformers 4.35.2
+- Pytorch 2.1.0a0+32f93b1
+- 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/mistral-7b-sft-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": [
+    "q_proj",
+    "v_proj",
+    "k_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac0ac612cfe9f6f0d063be9c4e6411fd08f9c319bc6d653e846e0daacf26ad2b
+size 109086672

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 0.99,
+    "eval_loss": 0.8835160732269287,
+    "eval_runtime": 41.2512,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 48.483,
+    "eval_steps_per_second": 0.776,
+    "train_loss": 0.9063981016476949,
+    "train_runtime": 3868.736,
+    "train_samples": 61966,
+    "train_samples_per_second": 16.017,
+    "train_steps_per_second": 0.004
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.99,
+    "eval_loss": 0.8835160732269287,
+    "eval_runtime": 41.2512,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 48.483,
+    "eval_steps_per_second": 0.776
+}

runs/Nov19_07-58-05_batch-p4d-cr1-002/events.out.tfevents.1700409731.batch-p4d-cr1-002.3839667.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b509ac6077fd0f20f2a333735644c8bad858a06227cbfaa82570f377ef9d4449
+size 4379

runs/Nov19_08-19-38_batch-p4d-cr1-007/events.out.tfevents.1700411219.batch-p4d-cr1-007.168213.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2591e12d05785d07990ba1217653411fa612664391e39e5c93dcefb4f5750b2
+size 5455

runs/Nov19_08-19-38_batch-p4d-cr1-007/events.out.tfevents.1700415129.batch-p4d-cr1-007.168213.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0edf9f23928481a7778603e8366e9de51d9685539f07ab5e7ae4f4e41f0d60bb
+size 306

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": 0.99,
+    "train_loss": 0.9063981016476949,
+    "train_runtime": 3868.736,
+    "train_samples": 61966,
+    "train_samples_per_second": 16.017,
+    "train_steps_per_second": 0.004
+}

trainer_state.json

@@ -0,0 +1,60 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9912235415591121,
+  "eval_steps": 500,
+  "global_step": 15,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9781476007338058e-05,
+      "loss": 0.9148,
+      "step": 1
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.9129,
+      "step": 5
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 5.000000000000003e-06,
+      "loss": 0.906,
+      "step": 10
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.0,
+      "loss": 0.8999,
+      "step": 15
+    },
+    {
+      "epoch": 0.99,
+      "eval_loss": 0.8835160732269287,
+      "eval_runtime": 42.1009,
+      "eval_samples_per_second": 47.505,
+      "eval_steps_per_second": 0.76,
+      "step": 15
+    },
+    {
+      "epoch": 0.99,
+      "step": 15,
+      "total_flos": 1.2602014135484416e+16,
+      "train_loss": 0.9063981016476949,
+      "train_runtime": 3868.736,
+      "train_samples_per_second": 16.017,
+      "train_steps_per_second": 0.004
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 15,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 1.2602014135484416e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e14de8fe9242bd33c4eb6af3ce7c405ef684a9d53d19236bc4369ffd2b374d38
+size 5624