Downloads · 30 days
237
2% of all-time downloads
RichardErkhov/ruslandev_-_llama-3-8b-samantha-gguf
ruslandev_-_llama-3-8b-samantha-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
237
2% of all-time downloads
All-time downloads
9.7K
Public
Repo size
92.9 GB
Likes
0
Public
Click a slice to open those files.
.gguf108 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
llama-3-8b-samantha - GGUF
| Name | Quant method | Size |
|---|---|---|
| llama-3-8b-samantha.Q2_K.gguf | Q2_K | 2.96GB |
| llama-3-8b-samantha.IQ3_XS.gguf | IQ3_XS | 3.28GB |
| llama-3-8b-samantha.IQ3_S.gguf | IQ3_S | 3.43GB |
| llama-3-8b-samantha.Q3_K_S.gguf | Q3_K_S | 3.41GB |
| llama-3-8b-samantha.IQ3_M.gguf | IQ3_M | 3.52GB |
| llama-3-8b-samantha.Q3_K.gguf | Q3_K | 3.74GB |
| llama-3-8b-samantha.Q3_K_M.gguf | Q3_K_M | 3.74GB |
| llama-3-8b-samantha.Q3_K_L.gguf | Q3_K_L | 4.03GB |
| llama-3-8b-samantha.IQ4_XS.gguf | IQ4_XS | 4.18GB |
| llama-3-8b-samantha.Q4_0.gguf | Q4_0 | 4.34GB |
| llama-3-8b-samantha.IQ4_NL.gguf | IQ4_NL | 4.38GB |
| llama-3-8b-samantha.Q4_K_S.gguf | Q4_K_S | 4.37GB |
| llama-3-8b-samantha.Q4_K.gguf | Q4_K | 4.58GB |
| llama-3-8b-samantha.Q4_K_M.gguf | Q4_K_M | 4.58GB |
| llama-3-8b-samantha.Q4_1.gguf | Q4_1 | 4.78GB |
| llama-3-8b-samantha.Q5_0.gguf | Q5_0 | 5.21GB |
| llama-3-8b-samantha.Q5_K_S.gguf | Q5_K_S | 5.21GB |
| llama-3-8b-samantha.Q5_K.gguf | Q5_K | 5.34GB |
| llama-3-8b-samantha.Q5_K_M.gguf | Q5_K_M | 5.34GB |
| llama-3-8b-samantha.Q5_1.gguf | Q5_1 | 5.65GB |
| llama-3-8b-samantha.Q6_K.gguf | Q6_K | 6.14GB |
| llama-3-8b-samantha.Q8_0.gguf | Q8_0 | 7.95GB |
language:
This model is finetuned on the data of Samantha.
Prompt format is Alpaca. I used the same system prompt as the original Samantha.
"""Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
### Instruction:
{SYSTEM_PROMPT}
### Input:
{QUESTION}
### Response:
"""
gptchain framework has been used for training.
| Training Loss | Epoch | Step |
|---|---|---|
| 2.0778 | 0.0 | 1 |
| 0.6255 | 0.18 | 120 |
| 0.6208 | 0.94 | 620 |
| 0.6244 | 2.0 | 1306 |
2 epoch finetuning from llama-3-8b took 1 hour on a single A100 with Unsloth and Huggingface's TRL library.