Update README.md

README.md

@@ -1,3 +1,97 @@
----
-license: llama3
----
+---
+base_model:
+- elinas/Llama-3-13B-Instruct
+library_name: transformers
+tags:
+- mergekit
+- merge
+license: llama3
+---
+# Meta-Llama-3-13B-Instruct
+
+This is a QLoRA **finetune** of a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
+
+The model is based on my passthrough merge of [Llama-3-13B-Instruct](https://huggingface.co/elinas/Llama-3-13B-Instruct)
+
+This was primarily an experiment to see how a passthrough merge will respond to further finetuning, though this was done on a small dataset.
+
+The model was finetuned on 8192 context length and is likely reliable using RoPE scaling up to 32k.
+
+It still cannot do math reliably; neither can Llama-3-8B, and in my tests only Llama-3-70B can, but it a better storywriter/RP than Llama-3-8B from some side by side testing.
+
+## Dataset
+
+* Dataset used
+   * [Chat-Error/Pure-dove-sharegpt](https://huggingface.co/datasets/Chat-Error/Pure-dove-sharegpt)
+
+A small dataset was used to see how it affects performance. Originally I planned to do a larger dataset (196k samples), but wanted to start with a smaller one first to see how much the model improved with some additional finetuning. 
+Next steps would be finetuning on a larger dataset if through further testing, performance improvements are noticed.
+
+## Finetuning details
+This is a QLoRA model and all modules were targeted.
+```yaml
+lora_target_modules:
+  - gate_proj
+  - down_proj
+  - up_proj
+  - q_proj
+  - v_proj
+  - k_proj
+  - o_proj
+lora_modules_to_save:
+  - embed_tokens
+  - lm_head
+```
+
+```yaml
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 1
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 3
+- total_train_batch_size: 3
+- total_eval_batch_size: 3
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 25
+- num_epochs: 1
+```
+
+Optimizer `paged_adamw_8bit` and Deepspeed ZeRO 3 was used at a LR of `1e-5` using the cosine scheduler for 1 epoch on 3x3090s taking 4h 12m 13s total.
+
+Sample packing and padding was disabled to reduce VRAM consumption significantly at the cost of speed.
+
+W&B Run Summary
+```
+wandb: Run summary:
+wandb:                eval/loss 1.00774
+wandb:             eval/runtime 535.3847
+wandb:  eval/samples_per_second 0.721
+wandb:    eval/steps_per_second 0.241
+wandb:               total_flos 4167452590080.0
+wandb:              train/epoch 1.0
+wandb:        train/global_step 1157
+wandb:          train/grad_norm 4.50846
+wandb:      train/learning_rate 0.0
+wandb:               train/loss 1.4115
+wandb:               train_loss 1.00352
+wandb:            train_runtime 14921.1227
+wandb: train_samples_per_second 0.233
+wandb:   train_steps_per_second 0.078
+```
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.0.dev0
+- Pytorch 2.3.0+cu121
+- Datasets 2.15.0
+- Tokenizers 0.15.0
+
+## Evaluations
+
+TBD - submitted
+
+[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)