Model save

README.md

@@ -1,57 +1,64 @@
 ---
 base_model: facebook/opt-350m
-library_name: transformers
-model_name: opt350
+datasets:
+- generator
+library_name: peft
+license: other
 tags:
-- generated_from_trainer
 - trl
 - sft
-licence: license
+- generated_from_trainer
+model-index:
+- name: opt350
+  results: []
 ---
 
-# Model Card for opt350
+<!-- 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. -->
 
-This model is a fine-tuned version of [facebook/opt-350m](https://huggingface.co/facebook/opt-350m).
-It has been trained using [TRL](https://github.com/huggingface/trl).
+# opt350
 
-## Quick start
+This model is a fine-tuned version of [facebook/opt-350m](https://huggingface.co/facebook/opt-350m) on the generator dataset.
 
-```python
-from transformers import pipeline
+## Model description
 
-question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
-generator = pipeline("text-generation", model="zhangduo4610/opt350", device="cuda")
-output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
-print(output["generated_text"])
-```
+More information needed
 
-## Training procedure
+## Intended uses & limitations
 
+More information needed
 
+## Training and evaluation data
 
-This model was trained with SFT.
+More information needed
 
-### Framework versions
+## Training procedure
+
+### Training hyperparameters
 
-- TRL: 0.12.0
-- Transformers: 4.46.1
-- Pytorch: 2.1.2
-- Datasets: 3.1.0
-- Tokenizers: 0.20.2
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- total_eval_batch_size: 16
+- 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
 
-## Citations
+### Training results
 
 
 
-Cite TRL as:
-    
-```bibtex
-@misc{vonwerra2022trl,
-	title        = {{TRL: Transformer Reinforcement Learning}},
-	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
-	year         = 2020,
-	journal      = {GitHub repository},
-	publisher    = {GitHub},
-	howpublished = {\url{https://github.com/huggingface/trl}}
-}
-```
+### Framework versions
+
+- PEFT 0.13.2
+- Transformers 4.45.0
+- Pytorch 2.1.2
+- Datasets 3.1.0
+- Tokenizers 0.20.2

adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "q_proj",
     "k_proj",
-    "gate_proj",
+    "v_proj",
+    "up_proj",
     "o_proj",
-    "down_proj",
-    "up_proj"
+    "gate_proj",
+    "q_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.9999380306128772,
     "total_flos": 4248659495485440.0,
-    "train_loss": 0.39469828721492317,
-    "train_runtime": 1041.1882,
+    "train_loss": 0.0,
+    "train_runtime": 0.0126,
     "train_samples": 207864,
-    "train_samples_per_second": 123.988,
-    "train_steps_per_second": 7.749
+    "train_samples_per_second": 10236188.723,
+    "train_steps_per_second": 639727.105
 }

runs/Nov05_10-32-53_gnode007.cluster/events.out.tfevents.1730832694.gnode007.cluster.1851533.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4e10a62d6bf762c2dc341c44ec2155e7d91b27293b2fc50ab064f4039655a64
+size 5968

runs/Nov05_10-54-00_gnode007.cluster/events.out.tfevents.1730832849.gnode007.cluster.1852419.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3a39a1bf8021df3f3214325d182c14dde90e0867dcca30d2cef787bbb319172
+size 5968

tokenizer_config.json

@@ -21,7 +21,7 @@
   },
   "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,
+  "clean_up_tokenization_spaces": true,
   "eos_token": "</s>",
   "errors": "replace",
   "model_max_length": 2048,

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.9999380306128772,
     "total_flos": 4248659495485440.0,
-    "train_loss": 0.39469828721492317,
-    "train_runtime": 1041.1882,
+    "train_loss": 0.0,
+    "train_runtime": 0.0126,
     "train_samples": 207864,
-    "train_samples_per_second": 123.988,
-    "train_steps_per_second": 7.749
+    "train_samples_per_second": 10236188.723,
+    "train_steps_per_second": 639727.105
 }

trainer_state.json

@@ -11306,22 +11306,14 @@
       "loss": 3.5602,
       "step": 8065
     },
-    {
-      "epoch": 0.9999380306128772,
-      "eval_loss": 1.7876207828521729,
-      "eval_runtime": 155.4034,
-      "eval_samples_per_second": 91.883,
-      "eval_steps_per_second": 5.746,
-      "step": 8068
-    },
     {
       "epoch": 0.9999380306128772,
       "step": 8068,
       "total_flos": 4248659495485440.0,
-      "train_loss": 0.39469828721492317,
-      "train_runtime": 1041.1882,
-      "train_samples_per_second": 123.988,
-      "train_steps_per_second": 7.749
+      "train_loss": 0.0,
+      "train_runtime": 0.0126,
+      "train_samples_per_second": 10236188.723,
+      "train_steps_per_second": 639727.105
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b40232365cdac60e63e903629fe55e11c030e9274cbe1a3765db331b581f516c
-size 7096
+oid sha256:258e43470debd52839f53189bb0668179eb156a1bcda2b08fa180bb781f216cc
+size 7032