Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,93 @@
+---
+base_model: habanoz/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1
+datasets:
+- OpenAssistant/oasst_top1_2023-08-25
+inference: false
+language:
+- en
+license: apache-2.0
+model_creator: habanoz
+model_name: TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1
+pipeline_tag: text-generation
+quantized_by: afrideva
+tags:
+- gguf
+- ggml
+- quantized
+- q2_k
+- q3_k_m
+- q4_k_m
+- q5_k_m
+- q6_k
+- q8_0
+---
+# habanoz/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1-GGUF
+
+Quantized GGUF model files for [TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1](https://huggingface.co/habanoz/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1) from [habanoz](https://huggingface.co/habanoz)
+
+
+| Name | Quant method | Size |
+| ---- | ---- | ---- |
+| [tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.fp16.gguf](https://huggingface.co/afrideva/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1-GGUF/resolve/main/tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.fp16.gguf) | fp16 | 2.20 GB  |
+| [tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q2_k.gguf](https://huggingface.co/afrideva/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1-GGUF/resolve/main/tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q2_k.gguf) | q2_k | 483.12 MB  |
+| [tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q3_k_m.gguf](https://huggingface.co/afrideva/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1-GGUF/resolve/main/tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q3_k_m.gguf) | q3_k_m | 550.82 MB  |
+| [tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q4_k_m.gguf](https://huggingface.co/afrideva/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1-GGUF/resolve/main/tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q4_k_m.gguf) | q4_k_m | 668.79 MB  |
+| [tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q5_k_m.gguf](https://huggingface.co/afrideva/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1-GGUF/resolve/main/tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q5_k_m.gguf) | q5_k_m | 783.02 MB  |
+| [tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q6_k.gguf](https://huggingface.co/afrideva/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1-GGUF/resolve/main/tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q6_k.gguf) | q6_k | 904.39 MB  |
+| [tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q8_0.gguf](https://huggingface.co/afrideva/TinyLlama-1.1B-step-2T-lr-5-5ep-oasst1-top1-instruct-V1-GGUF/resolve/main/tinyllama-1.1b-step-2t-lr-5-5ep-oasst1-top1-instruct-v1.q8_0.gguf) | q8_0 | 1.17 GB  |
+
+
+
+## Original Model Card:
+TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T finetuned using OpenAssistant/oasst_top1_2023-08-25 dataset. 
+
+Trained for 5 epochs using Qlora. Adapter is merged.
+
+SFT code:
+https://github.com/habanoz/qlora.git
+
+Command used:
+```bash
+accelerate launch $BASE_DIR/qlora/train.py \
+  --model_name_or_path $BASE_MODEL \
+  --working_dir $BASE_DIR/$OUTPUT_NAME-checkpoints \
+  --output_dir $BASE_DIR/$OUTPUT_NAME-peft \
+  --merged_output_dir $BASE_DIR/$OUTPUT_NAME \
+  --final_output_dir $BASE_DIR/$OUTPUT_NAME-final \
+  --num_train_epochs 5 \
+  --logging_steps 1 \
+  --save_strategy steps \
+  --save_steps 75 \
+  --save_total_limit 2 \
+  --data_seed 11422 \
+  --evaluation_strategy steps \
+  --per_device_eval_batch_size 4 \
+  --eval_dataset_size 0.01 \
+  --eval_steps 75 \
+  --max_new_tokens 1024 \
+  --dataloader_num_workers 3 \
+  --logging_strategy steps \
+  --do_train \
+  --do_eval \
+  --lora_r 64 \
+  --lora_alpha 16 \
+  --lora_modules all \
+  --bits 4 \
+  --double_quant \
+  --quant_type nf4 \
+  --lr_scheduler_type constant \
+  --dataset oasst1-top1 \
+  --dataset_format oasst1 \
+  --model_max_len 1024 \
+  --per_device_train_batch_size 4 \
+  --gradient_accumulation_steps 4 \
+  --learning_rate 1e-5 \
+  --adam_beta2 0.999 \
+  --max_grad_norm 0.3 \
+  --lora_dropout 0.0 \
+  --weight_decay 0.0 \
+  --seed 11422 \
+  --gradient_checkpointing \
+  --use_flash_attention_2 \
+  --ddp_find_unused_parameters False
+```