114 lines
6.9 KiB
Markdown
114 lines
6.9 KiB
Markdown
|
Quantization made by Richard Erkhov.
|
||
|
|
|
||
|
|
[Github](https://github.com/RichardErkhov)
|
||
|
|
|
||
|
|
[Discord](https://discord.gg/pvy7H8DZMG)
|
||
|
|
|
||
|
|
[Request more models](https://github.com/RichardErkhov/quant_request)
|
||
|
|
|
||
|
|
|
||
|
|
llama3-1_8b_oh_v3.1_wo_share_gpt - GGUF
|
||
|
|
- Model creator: https://huggingface.co/mlfoundations-dev/
|
||
|
|
- Original model: https://huggingface.co/mlfoundations-dev/llama3-1_8b_oh_v3.1_wo_share_gpt/
|
||
|
|
|
||
|
|
|
||
|
|
| Name | Quant method | Size |
|
||
|
|
| ---- | ---- | ---- |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q2_K.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q2_K.gguf) | Q2_K | 2.96GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.IQ3_XS.gguf) | IQ3_XS | 3.28GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.IQ3_S.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.IQ3_S.gguf) | IQ3_S | 3.43GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.IQ3_M.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.IQ3_M.gguf) | IQ3_M | 3.52GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q3_K.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q3_K.gguf) | Q3_K | 3.74GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.IQ4_XS.gguf) | IQ4_XS | 4.18GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_0.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_0.gguf) | Q4_0 | 4.34GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.IQ4_NL.gguf) | IQ4_NL | 4.38GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_K_S.gguf) | Q4_K_S | 4.37GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_K.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_K.gguf) | Q4_K | 4.58GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_K_M.gguf) | Q4_K_M | 4.58GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_1.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q4_1.gguf) | Q4_1 | 4.78GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_0.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_0.gguf) | Q5_0 | 5.21GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_K.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_K.gguf) | Q5_K | 5.34GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_K_M.gguf) | Q5_K_M | 5.34GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_1.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q5_1.gguf) | Q5_1 | 5.65GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q6_K.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q6_K.gguf) | Q6_K | 6.14GB |
|
||
|
|
| [llama3-1_8b_oh_v3.1_wo_share_gpt.Q8_0.gguf](https://huggingface.co/RichardErkhov/mlfoundations-dev_-_llama3-1_8b_oh_v3.1_wo_share_gpt-gguf/blob/main/llama3-1_8b_oh_v3.1_wo_share_gpt.Q8_0.gguf) | Q8_0 | 7.95GB |
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
Original model description:
|
||
|
|
---
|
||
|
|
library_name: transformers
|
||
|
|
license: llama3.1
|
||
|
|
base_model: meta-llama/Meta-Llama-3.1-8B
|
||
|
|
tags:
|
||
|
|
- llama-factory
|
||
|
|
- full
|
||
|
|
- generated_from_trainer
|
||
|
|
model-index:
|
||
|
|
- name: llama3-1_8b_oh_v3.1_wo_share_gpt
|
||
|
|
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. -->
|
||
|
|
|
||
|
|
# llama3-1_8b_oh_v3.1_wo_share_gpt
|
||
|
|
|
||
|
|
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/oh_v3.1_wo_share_gpt dataset.
|
||
|
|
It achieves the following results on the evaluation set:
|
||
|
|
- Loss: 0.6453
|
||
|
|
|
||
|
|
## 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: 5e-06
|
||
|
|
- train_batch_size: 16
|
||
|
|
- eval_batch_size: 8
|
||
|
|
- seed: 42
|
||
|
|
- distributed_type: multi-GPU
|
||
|
|
- num_devices: 32
|
||
|
|
- total_train_batch_size: 512
|
||
|
|
- total_eval_batch_size: 256
|
||
|
|
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
||
|
|
- lr_scheduler_type: constant
|
||
|
|
- lr_scheduler_warmup_ratio: 0.1
|
||
|
|
- lr_scheduler_warmup_steps: 1738
|
||
|
|
- num_epochs: 3.0
|
||
|
|
|
||
|
|
### Training results
|
||
|
|
|
||
|
|
| Training Loss | Epoch | Step | Validation Loss |
|
||
|
|
|:-------------:|:-----:|:----:|:---------------:|
|
||
|
|
| 0.6517 | 1.0 | 422 | 0.6541 |
|
||
|
|
| 0.6045 | 2.0 | 844 | 0.6440 |
|
||
|
|
| 0.5731 | 3.0 | 1266 | 0.6453 |
|
||
|
|
|
||
|
|
|
||
|
|
### Framework versions
|
||
|
|
|
||
|
|
- Transformers 4.46.1
|
||
|
|
- Pytorch 2.4.0
|
||
|
|
- Datasets 3.0.2
|
||
|
|
- Tokenizers 0.20.3
|
||
|
|
|
||
|
|
|