| Quantization made by Richard Erkhov. | |
| [Github](https://github.com/RichardErkhov) | |
| [Discord](https://discord.gg/pvy7H8DZMG) | |
| [Request more models](https://github.com/RichardErkhov/quant_request) | |
| llama3-8b-math-sft-subtask-7 - GGUF | |
| - Model creator: https://huggingface.co/Dynosaur/ | |
| - Original model: https://huggingface.co/Dynosaur/llama3-8b-math-sft-subtask-7/ | |
| | Name | Quant method | Size | | |
| | ---- | ---- | ---- | | |
| | [llama3-8b-math-sft-subtask-7.Q2_K.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q2_K.gguf) | Q2_K | 2.96GB | | |
| | [llama3-8b-math-sft-subtask-7.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.IQ3_XS.gguf) | IQ3_XS | 3.28GB | | |
| | [llama3-8b-math-sft-subtask-7.IQ3_S.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.IQ3_S.gguf) | IQ3_S | 3.43GB | | |
| | [llama3-8b-math-sft-subtask-7.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q3_K_S.gguf) | Q3_K_S | 3.41GB | | |
| | [llama3-8b-math-sft-subtask-7.IQ3_M.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.IQ3_M.gguf) | IQ3_M | 3.52GB | | |
| | [llama3-8b-math-sft-subtask-7.Q3_K.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q3_K.gguf) | Q3_K | 3.74GB | | |
| | [llama3-8b-math-sft-subtask-7.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q3_K_M.gguf) | Q3_K_M | 3.74GB | | |
| | [llama3-8b-math-sft-subtask-7.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q3_K_L.gguf) | Q3_K_L | 4.03GB | | |
| | [llama3-8b-math-sft-subtask-7.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.IQ4_XS.gguf) | IQ4_XS | 4.18GB | | |
| | [llama3-8b-math-sft-subtask-7.Q4_0.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q4_0.gguf) | Q4_0 | 4.34GB | | |
| | [llama3-8b-math-sft-subtask-7.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.IQ4_NL.gguf) | IQ4_NL | 4.38GB | | |
| | [llama3-8b-math-sft-subtask-7.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q4_K_S.gguf) | Q4_K_S | 4.37GB | | |
| | [llama3-8b-math-sft-subtask-7.Q4_K.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q4_K.gguf) | Q4_K | 4.58GB | | |
| | [llama3-8b-math-sft-subtask-7.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q4_K_M.gguf) | Q4_K_M | 4.58GB | | |
| | [llama3-8b-math-sft-subtask-7.Q4_1.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q4_1.gguf) | Q4_1 | 4.78GB | | |
| | [llama3-8b-math-sft-subtask-7.Q5_0.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q5_0.gguf) | Q5_0 | 5.21GB | | |
| | [llama3-8b-math-sft-subtask-7.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q5_K_S.gguf) | Q5_K_S | 5.21GB | | |
| | [llama3-8b-math-sft-subtask-7.Q5_K.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q5_K.gguf) | Q5_K | 5.34GB | | |
| | [llama3-8b-math-sft-subtask-7.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q5_K_M.gguf) | Q5_K_M | 5.34GB | | |
| | [llama3-8b-math-sft-subtask-7.Q5_1.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q5_1.gguf) | Q5_1 | 5.65GB | | |
| | [llama3-8b-math-sft-subtask-7.Q6_K.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q6_K.gguf) | Q6_K | 6.14GB | | |
| | [llama3-8b-math-sft-subtask-7.Q8_0.gguf](https://huggingface.co/RichardErkhov/Dynosaur_-_llama3-8b-math-sft-subtask-7-gguf/blob/main/llama3-8b-math-sft-subtask-7.Q8_0.gguf) | Q8_0 | 7.95GB | | |
| Original model description: | |
| --- | |
| library_name: transformers | |
| license: llama3 | |
| base_model: Dynosaur/llama3-8b-math-sft | |
| tags: | |
| - alignment-handbook | |
| - trl | |
| - sft | |
| - generated_from_trainer | |
| - trl | |
| - sft | |
| - generated_from_trainer | |
| datasets: | |
| - Dynosaur/math-sft-subtask-7 | |
| model-index: | |
| - name: llama3-8b-math-sft-subtask-7 | |
| 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. --> | |
| # llama3-8b-math-sft-subtask-7 | |
| This model is a fine-tuned version of [Dynosaur/llama3-8b-math-sft](https://huggingface.co/Dynosaur/llama3-8b-math-sft) on the Dynosaur/math-sft-subtask-7 dataset. | |
| ## 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: 2e-05 | |
| - train_batch_size: 2 | |
| - eval_batch_size: 8 | |
| - seed: 42 | |
| - distributed_type: multi-GPU | |
| - num_devices: 8 | |
| - gradient_accumulation_steps: 8 | |
| - total_train_batch_size: 128 | |
| - 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: 2 | |
| ### Training results | |
| ### Framework versions | |
| - Transformers 4.44.2 | |
| - Pytorch 2.4.1+cu121 | |
| - Datasets 3.0.0 | |
| - Tokenizers 0.19.1 | |