<< YOUR USER NAME HERE>>/llama381binstruct_summarize_short_updated

README.md

@@ -1,83 +1,58 @@
 ---
 base_model: NousResearch/Meta-Llama-3.1-8B-Instruct
-datasets:
-- generator
-library_name: peft
-license: llama3.1
+library_name: transformers
+model_name: llama381binstruct_summarize_short
 tags:
+- generated_from_trainer
 - trl
 - sft
-- generated_from_trainer
-model-index:
-- name: llama381binstruct_summarize_short
-  results: []
+licence: license
 ---
 
-<!-- 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. -->
-
-# llama381binstruct_summarize_short
-
-This model is a fine-tuned version of [NousResearch/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/NousResearch/Meta-Llama-3.1-8B-Instruct) on the generator dataset.
-It achieves the following results on the evaluation set:
-- Loss: 1.8652
+# Model Card for llama381binstruct_summarize_short
 
-## Model description
+This model is a fine-tuned version of [NousResearch/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/NousResearch/Meta-Llama-3.1-8B-Instruct).
+It has been trained using [TRL](https://github.com/huggingface/trl).
 
-More information needed
+## Quick start
 
-## Intended uses & limitations
+```python
+from transformers import pipeline
 
-More information needed
+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="chenshi910814/llama381binstruct_summarize_short", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
 
-## Training and evaluation data
+## Training procedure
 
-More information needed
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/chenshi910814-northern-arizona-university/huggingface/runs/w09ae3dd) 
 
-## Training procedure
 
-### Training hyperparameters
+This model was trained with SFT.
 
-The following hyperparameters were used during training:
-- learning_rate: 0.0002
-- train_batch_size: 1
-- eval_batch_size: 8
-- seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 30
-- training_steps: 500
+### Framework versions
 
-### Training results
+- TRL: 0.16.0
+- Transformers: 4.50.3
+- Pytorch: 2.6.0+cu124
+- Datasets: 3.5.0
+- Tokenizers: 0.21.1
 
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| 1.4098        | 2.5   | 25   | 1.0767          |
-| 0.3872        | 5.0   | 50   | 1.2700          |
-| 0.0902        | 7.5   | 75   | 1.6628          |
-| 0.0294        | 10.0  | 100  | 1.5043          |
-| 0.0147        | 12.5  | 125  | 1.6301          |
-| 0.007         | 15.0  | 150  | 1.6782          |
-| 0.005         | 17.5  | 175  | 1.7262          |
-| 0.0026        | 20.0  | 200  | 1.7412          |
-| 0.0013        | 22.5  | 225  | 1.7683          |
-| 0.0009        | 25.0  | 250  | 1.8120          |
-| 0.0008        | 27.5  | 275  | 1.8294          |
-| 0.0008        | 30.0  | 300  | 1.8397          |
-| 0.0007        | 32.5  | 325  | 1.8466          |
-| 0.0007        | 35.0  | 350  | 1.8521          |
-| 0.0006        | 37.5  | 375  | 1.8566          |
-| 0.0006        | 40.0  | 400  | 1.8599          |
-| 0.0006        | 42.5  | 425  | 1.8625          |
-| 0.0006        | 45.0  | 450  | 1.8641          |
-| 0.0006        | 47.5  | 475  | 1.8653          |
-| 0.0005        | 50.0  | 500  | 1.8652          |
+## Citations
 
 
-### Framework versions
 
-- PEFT 0.13.2
-- Transformers 4.45.2
-- Pytorch 2.4.1+cu121
-- Datasets 3.0.1
-- Tokenizers 0.20.1
+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}}
+}
+```

adapter_config.json

@@ -3,6 +3,9 @@
   "auto_mapping": null,
   "base_model_name_or_path": "NousResearch/Meta-Llama-3.1-8B-Instruct",
   "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
   "fan_in_fan_out": false,
   "inference_mode": true,
   "init_lora_weights": true,
@@ -11,6 +14,7 @@
   "layers_to_transform": null,
   "loftq_config": {},
   "lora_alpha": 32,
+  "lora_bias": false,
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
@@ -20,15 +24,16 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
-    "v_proj",
     "down_proj",
+    "v_proj",
+    "gate_proj",
     "o_proj",
-    "q_proj",
     "k_proj",
+    "q_proj",
     "up_proj"
   ],
   "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
   "use_dora": false,
   "use_rslora": false
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60d95b10b6e140a9626a7058d5038528f2ff80148dc4569b881db56052046509
-size 40
+oid sha256:5d7d362a843acdf447a18d8ffb7b832093a06f149a32d8f9e3c06fb000e06b7e
+size 167832240

runs/Apr01_15-58-02_13301a165851/events.out.tfevents.1743523279.13301a165851.225.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf57f648613d6d64c316d3e435be870457b2103264d5342ec475e8267151355f
+size 29690

tokenizer_config.json

@@ -2053,11 +2053,12 @@
   "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}",
   "clean_up_tokenization_spaces": true,
   "eos_token": "<|eot_id|>",
+  "extra_special_tokens": {},
   "model_input_names": [
     "input_ids",
     "attention_mask"
   ],
   "model_max_length": 131072,
   "pad_token": "<|eot_id|>",
-  "tokenizer_class": "PreTrainedTokenizerFast"
+  "tokenizer_class": "PreTrainedTokenizer"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:589437e567974d8a0a843e35793cbc6225174a09a1e8376ab2c9b5fe31e727a8
-size 5560
+oid sha256:8def6783973a11fe35efc1a9396dd1cdcd164474dff9588cd4d8ac26aa633b32
+size 5688