Model save

README.md

@@ -0,0 +1,72 @@
+---
+license: apache-2.0
+library_name: peft
+tags:
+- trl
+- dpo
+- generated_from_trainer
+base_model: sablo/sablo-pebble-mistral
+model-index:
+- name: sablo-pebble-mistral-dpo-lora-oasst2_dpo_pairs_en
+  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. -->
+
+# sablo-pebble-mistral-dpo-lora-oasst2_dpo_pairs_en
+
+This model is a fine-tuned version of [sablo/sablo-pebble-mistral](https://huggingface.co/sablo/sablo-pebble-mistral) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6714
+- Rewards/chosen: 0.1838
+- Rewards/rejected: 0.0314
+- Rewards/accuracies: 0.6875
+- Rewards/margins: 0.1524
+- Logps/rejected: -267.5412
+- Logps/chosen: -312.9823
+- Logits/rejected: -2.2363
+- Logits/chosen: -2.2654
+
+## 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-06
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.7.1
+- Transformers 4.36.2
+- Pytorch 2.1.2+cu118
+- Datasets 2.14.6
+- Tokenizers 0.15.1

adapter_config.json

@@ -0,0 +1,31 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "sablo/sablo-pebble-mistral",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_proj",
+    "o_proj",
+    "k_proj",
+    "q_proj",
+    "down_proj",
+    "v_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8b0a251b61db24c85bfa953214ad371e6dd568821e075977e969a2cc7b7e02f
+size 671150064

added_tokens.json

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

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 1.0,
+    "eval_logits/chosen": -2.265434980392456,
+    "eval_logits/rejected": -2.236332654953003,
+    "eval_logps/chosen": -312.9822692871094,
+    "eval_logps/rejected": -267.5411682128906,
+    "eval_loss": 0.6714204549789429,
+    "eval_rewards/accuracies": 0.6875,
+    "eval_rewards/chosen": 0.1837889701128006,
+    "eval_rewards/margins": 0.15236972272396088,
+    "eval_rewards/rejected": 0.031419236212968826,
+    "eval_runtime": 72.2288,
+    "eval_samples": 256,
+    "eval_samples_per_second": 3.544,
+    "eval_steps_per_second": 0.055,
+    "train_loss": 0.6752520862378573,
+    "train_runtime": 2732.0836,
+    "train_samples": 4860,
+    "train_samples_per_second": 1.779,
+    "train_steps_per_second": 0.028
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_logits/chosen": -2.265434980392456,
+    "eval_logits/rejected": -2.236332654953003,
+    "eval_logps/chosen": -312.9822692871094,
+    "eval_logps/rejected": -267.5411682128906,
+    "eval_loss": 0.6714204549789429,
+    "eval_rewards/accuracies": 0.6875,
+    "eval_rewards/chosen": 0.1837889701128006,
+    "eval_rewards/margins": 0.15236972272396088,
+    "eval_rewards/rejected": 0.031419236212968826,
+    "eval_runtime": 72.2288,
+    "eval_samples": 256,
+    "eval_samples_per_second": 3.544,
+    "eval_steps_per_second": 0.055
+}

runs/Feb08_12-57-41_rv-ai1/events.out.tfevents.1707397108.rv-ai1.4285.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72828c0ce659f2ce79fbf375a74009cf8f1ccca782e7f3908e23150cb3b8094c
+size 5658

runs/Feb08_13-04-37_rv-ai1/events.out.tfevents.1707397531.rv-ai1.11994.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03f9097cbfdbbd1c0059887090acdbf194474ca3004f3d61272892c534c0ba32
+size 9979

runs/Feb08_13-04-37_rv-ai1/events.out.tfevents.1707400336.rv-ai1.11994.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21d08fa81a9b44ed79b5a63c6c9cff48e3d24ee5fb5f3aff548526cc3f510feb
+size 815

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "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 +1,61 @@
+{
+  "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
+    },
+    "32000": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [],
+  "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": "<|im_end|>",
+  "legacy": true,
+  "model_max_length": 2048,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "trust_remote_code": false,
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false,
+  "use_fast": true
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 0.6752520862378573,
+    "train_runtime": 2732.0836,
+    "train_samples": 4860,
+    "train_samples_per_second": 1.779,
+    "train_steps_per_second": 0.028
+}

trainer_state.json

@@ -0,0 +1,142 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 100,
+  "global_step": 76,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.25e-07,
+      "logits/chosen": -2.521343946456909,
+      "logits/rejected": -2.3620429039001465,
+      "logps/chosen": -521.3419189453125,
+      "logps/rejected": -476.025146484375,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.989335440737587e-06,
+      "logits/chosen": -2.358004093170166,
+      "logits/rejected": -2.4325995445251465,
+      "logps/chosen": -345.5417175292969,
+      "logps/rejected": -309.4444580078125,
+      "loss": 0.6911,
+      "rewards/accuracies": 0.5694444179534912,
+      "rewards/chosen": 0.01685788854956627,
+      "rewards/margins": 0.004424349870532751,
+      "rewards/rejected": 0.012433539144694805,
+      "step": 10
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.625542839324036e-06,
+      "logits/chosen": -2.4209659099578857,
+      "logits/rejected": -2.4008195400238037,
+      "logps/chosen": -297.3746643066406,
+      "logps/rejected": -285.84234619140625,
+      "loss": 0.679,
+      "rewards/accuracies": 0.612500011920929,
+      "rewards/chosen": 0.07351277023553848,
+      "rewards/margins": 0.030746515840291977,
+      "rewards/rejected": 0.04276625066995621,
+      "step": 20
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.81608040719339e-06,
+      "logits/chosen": -2.4519193172454834,
+      "logits/rejected": -2.5038084983825684,
+      "logps/chosen": -297.0643005371094,
+      "logps/rejected": -273.5292053222656,
+      "loss": 0.6686,
+      "rewards/accuracies": 0.6625000238418579,
+      "rewards/chosen": 0.1535051167011261,
+      "rewards/margins": 0.07135292887687683,
+      "rewards/rejected": 0.08215219527482986,
+      "step": 30
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 2.730670898658255e-06,
+      "logits/chosen": -2.4741883277893066,
+      "logits/rejected": -2.297362804412842,
+      "logps/chosen": -321.7876281738281,
+      "logps/rejected": -261.57379150390625,
+      "loss": 0.673,
+      "rewards/accuracies": 0.699999988079071,
+      "rewards/chosen": 0.1949460208415985,
+      "rewards/margins": 0.07163376361131668,
+      "rewards/rejected": 0.12331227213144302,
+      "step": 40
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5968958345321178e-06,
+      "logits/chosen": -2.4305148124694824,
+      "logits/rejected": -2.4470670223236084,
+      "logps/chosen": -298.35980224609375,
+      "logps/rejected": -267.8696594238281,
+      "loss": 0.6744,
+      "rewards/accuracies": 0.6499999761581421,
+      "rewards/chosen": 0.09592735767364502,
+      "rewards/margins": 0.0405537411570549,
+      "rewards/rejected": 0.05537362024188042,
+      "step": 50
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 6.524777069483526e-07,
+      "logits/chosen": -2.4608042240142822,
+      "logits/rejected": -2.472360849380493,
+      "logps/chosen": -333.09967041015625,
+      "logps/rejected": -280.1986389160156,
+      "loss": 0.6734,
+      "rewards/accuracies": 0.737500011920929,
+      "rewards/chosen": 0.23129253089427948,
+      "rewards/margins": 0.10529744625091553,
+      "rewards/rejected": 0.12599506974220276,
+      "step": 60
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 9.54358920679524e-08,
+      "logits/chosen": -2.3924319744110107,
+      "logits/rejected": -2.393690586090088,
+      "logps/chosen": -306.0284729003906,
+      "logps/rejected": -271.36529541015625,
+      "loss": 0.6684,
+      "rewards/accuracies": 0.512499988079071,
+      "rewards/chosen": 0.06389348953962326,
+      "rewards/margins": 0.027882063761353493,
+      "rewards/rejected": 0.03601142764091492,
+      "step": 70
+    },
+    {
+      "epoch": 1.0,
+      "step": 76,
+      "total_flos": 0.0,
+      "train_loss": 0.6752520862378573,
+      "train_runtime": 2732.0836,
+      "train_samples_per_second": 1.779,
+      "train_steps_per_second": 0.028
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 76,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "total_flos": 0.0,
+  "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 sha256:21b0a30d45e44a9f8889106a8ad2870f49ddea159d0e5e301ca5f86d802ac07b
+size 6072