Downloads · 30 days
87
5% of all-time downloads
RichardErkhov/ewof_-_koishi-instruct-3b-gguf
ewof_-_koishi-instruct-3b-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
87
5% of all-time downloads
All-time downloads
1.9K
Public
Repo size
32.4 GB
Likes
0
Public
Click a slice to open those files.
.gguf37.8 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
koishi-instruct-3b - GGUF
| Name | Quant method | Size |
|---|---|---|
| koishi-instruct-3b.Q2_K.gguf | Q2_K | 1.01GB |
| koishi-instruct-3b.IQ3_XS.gguf | IQ3_XS | 1.14GB |
| koishi-instruct-3b.IQ3_S.gguf | IQ3_S | 1.16GB |
| koishi-instruct-3b.Q3_K_S.gguf | Q3_K_S | 1.16GB |
| koishi-instruct-3b.IQ3_M.gguf | IQ3_M | 1.28GB |
| koishi-instruct-3b.Q3_K.gguf | Q3_K | 1.38GB |
| koishi-instruct-3b.Q3_K_M.gguf | Q3_K_M | 1.38GB |
| koishi-instruct-3b.Q3_K_L.gguf | Q3_K_L | 1.49GB |
| koishi-instruct-3b.IQ4_XS.gguf | IQ4_XS | 1.43GB |
| koishi-instruct-3b.Q4_0.gguf | Q4_0 | 1.49GB |
| koishi-instruct-3b.IQ4_NL.gguf | IQ4_NL | 1.5GB |
| koishi-instruct-3b.Q4_K_S.gguf | Q4_K_S | 1.5GB |
| koishi-instruct-3b.Q4_K.gguf | Q4_K | 1.66GB |
| koishi-instruct-3b.Q4_K_M.gguf | Q4_K_M | 1.66GB |
| koishi-instruct-3b.Q4_1.gguf | Q4_1 | 1.64GB |
| koishi-instruct-3b.Q5_0.gguf | Q5_0 | 1.8GB |
| koishi-instruct-3b.Q5_K_S.gguf | Q5_K_S | 1.8GB |
| koishi-instruct-3b.Q5_K.gguf | Q5_K | 1.93GB |
| koishi-instruct-3b.Q5_K_M.gguf | Q5_K_M | 1.93GB |
| koishi-instruct-3b.Q5_1.gguf | Q5_1 | 1.95GB |
| koishi-instruct-3b.Q6_K.gguf | Q6_K | 2.13GB |
| koishi-instruct-3b.Q8_0.gguf | Q8_0 | 2.75GB |
datasets:
native fine tune of togethercomputer/RedPajama-INCITE-Base-3B-v1
The current model version has been trained on prompts using three different roles, which are denoted by the following tokens: <|system|>, <|user|> and <|model|>.
The <|system|> prompt can be used to inject out-of-channel information behind the scenes, while the <|user|> prompt should be used to indicate user input. The <|model|> token should then be used to indicate that the model should generate a response. These tokens can happen multiple times and be chained up to form a conversation history.
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 33.48 |
| ARC (25-shot) | 40.96 |
| HellaSwag (10-shot) | 64.54 |
| MMLU (5-shot) | 26.58 |
| TruthfulQA (0-shot) | 31.65 |
| Winogrande (5-shot) | 64.09 |
| GSM8K (5-shot) | 1.14 |
| DROP (3-shot) | 5.41 |