Upload 13 files

README.md

@@ -0,0 +1,62 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: IlyaGusev/saiga_mistral_7b_merged
+model-index:
+- name: sft
+  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
+
+This model is a fine-tuned version of [IlyaGusev/saiga_mistral_7b_merged](https://huggingface.co/IlyaGusev/saiga_mistral_7b_merged) on the gazeta dataset.
+
+## 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: 5e-05
+- train_batch_size: 4
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- total_eval_batch_size: 4
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 1.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.10.1.dev0
+- Transformers 4.39.1
+- Pytorch 2.1.0+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "IlyaGusev/saiga_mistral_7b_merged",
+  "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": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": "unsloth",
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "gate_proj",
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "down_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:b33ed03035a4d2bc0cefca53845e8440c0982a61b01f0d9cd18bdcad97c6ae31
+size 83945296

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "<|im_end|>": 32000,
+  "<|im_start|>": 32001
+}

all_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.0112789268493652,
+    "train_runtime": 5845.5599,
+    "train_samples_per_second": 0.342,
+    "train_steps_per_second": 0.021
+}

special_tokens_map.json

@@ -0,0 +1,35 @@
+{
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "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 +1,66 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32000": {
+      "content": "<|im_end|>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32001": {
+      "content": "<|im_start|>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>"
+  ],
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\\nAssistant: ' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' + '\\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 32768,
+  "pad_token": "<unk>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.0112789268493652,
+    "train_runtime": 5845.5599,
+    "train_samples_per_second": 0.342,
+    "train_steps_per_second": 0.021
+}

trainer_log.jsonl

@@ -0,0 +1,13 @@
+{"current_steps": 10, "total_steps": 125, "loss": 1.1006, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9214579028215776e-05, "epoch": 0.08, "percentage": 8.0, "elapsed_time": "0:07:39", "remaining_time": "1:28:05"}
+{"current_steps": 20, "total_steps": 125, "loss": 1.049, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.690766700109659e-05, "epoch": 0.16, "percentage": 16.0, "elapsed_time": "0:15:35", "remaining_time": "1:21:52"}
+{"current_steps": 30, "total_steps": 125, "loss": 1.0269, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.3224215685535294e-05, "epoch": 0.24, "percentage": 24.0, "elapsed_time": "0:23:16", "remaining_time": "1:13:41"}
+{"current_steps": 40, "total_steps": 125, "loss": 0.9896, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8395669874474915e-05, "epoch": 0.32, "percentage": 32.0, "elapsed_time": "0:30:51", "remaining_time": "1:05:35"}
+{"current_steps": 50, "total_steps": 125, "loss": 1.0179, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.272542485937369e-05, "epoch": 0.4, "percentage": 40.0, "elapsed_time": "0:38:41", "remaining_time": "0:58:01"}
+{"current_steps": 60, "total_steps": 125, "loss": 0.9636, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.656976298823284e-05, "epoch": 0.48, "percentage": 48.0, "elapsed_time": "0:46:22", "remaining_time": "0:50:14"}
+{"current_steps": 70, "total_steps": 125, "loss": 1.0151, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.031546713535688e-05, "epoch": 0.56, "percentage": 56.0, "elapsed_time": "0:53:53", "remaining_time": "0:42:20"}
+{"current_steps": 80, "total_steps": 125, "loss": 1.0039, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4355517710873184e-05, "epoch": 0.64, "percentage": 64.0, "elapsed_time": "1:01:41", "remaining_time": "0:34:42"}
+{"current_steps": 90, "total_steps": 125, "loss": 0.9773, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.064400256282757e-06, "epoch": 0.72, "percentage": 72.0, "elapsed_time": "1:09:26", "remaining_time": "0:27:00"}
+{"current_steps": 100, "total_steps": 125, "loss": 0.9578, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.7745751406263165e-06, "epoch": 0.8, "percentage": 80.0, "elapsed_time": "1:17:13", "remaining_time": "0:19:18"}
+{"current_steps": 110, "total_steps": 125, "loss": 0.9773, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7555878527937164e-06, "epoch": 0.88, "percentage": 88.0, "elapsed_time": "1:25:03", "remaining_time": "0:11:35"}
+{"current_steps": 120, "total_steps": 125, "loss": 1.0422, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9713246713805588e-07, "epoch": 0.96, "percentage": 96.0, "elapsed_time": "1:32:57", "remaining_time": "0:03:52"}
+{"current_steps": 125, "total_steps": 125, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "1:36:53", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,114 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 125,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.08,
+      "grad_norm": 1.7232568264007568,
+      "learning_rate": 4.9214579028215776e-05,
+      "loss": 1.1006,
+      "step": 10
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 2.3090097904205322,
+      "learning_rate": 4.690766700109659e-05,
+      "loss": 1.049,
+      "step": 20
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 1.6444264650344849,
+      "learning_rate": 4.3224215685535294e-05,
+      "loss": 1.0269,
+      "step": 30
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 1.920586109161377,
+      "learning_rate": 3.8395669874474915e-05,
+      "loss": 0.9896,
+      "step": 40
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 1.7457690238952637,
+      "learning_rate": 3.272542485937369e-05,
+      "loss": 1.0179,
+      "step": 50
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 1.9631178379058838,
+      "learning_rate": 2.656976298823284e-05,
+      "loss": 0.9636,
+      "step": 60
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 2.050771951675415,
+      "learning_rate": 2.031546713535688e-05,
+      "loss": 1.0151,
+      "step": 70
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 1.9484338760375977,
+      "learning_rate": 1.4355517710873184e-05,
+      "loss": 1.0039,
+      "step": 80
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 1.7718967199325562,
+      "learning_rate": 9.064400256282757e-06,
+      "loss": 0.9773,
+      "step": 90
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 2.00250506401062,
+      "learning_rate": 4.7745751406263165e-06,
+      "loss": 0.9578,
+      "step": 100
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 2.0536956787109375,
+      "learning_rate": 1.7555878527937164e-06,
+      "loss": 0.9773,
+      "step": 110
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 2.105520248413086,
+      "learning_rate": 1.9713246713805588e-07,
+      "loss": 1.0422,
+      "step": 120
+    },
+    {
+      "epoch": 1.0,
+      "step": 125,
+      "total_flos": 1.7032957304347034e+17,
+      "train_loss": 1.0112789268493652,
+      "train_runtime": 5845.5599,
+      "train_samples_per_second": 0.342,
+      "train_steps_per_second": 0.021
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 125,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1000,
+  "total_flos": 1.7032957304347034e+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 sha256:0a5374e5016f75dd7641625eb384725fff3af38fc71a5b14d2d6cbd148d8f66a
+size 5112