Downloads · 30 days
28
1% of all-time downloads
Ba2han_-_Tinypus-1.5B-gguf
RichardErkhov/Ba2han_-_Tinypus-1.5B-gguf
Ba2han_-_Tinypus-1.5B-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.
All-time downloads
4.4K
Public
Repo size
16.7 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf19.3 GB · 100%
At a glance
- Access
- Public
- Created
- Jun 25, 2024
- Updated
- Jun 25, 2024
- SHA
- 3bd1d8ff
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
Tinypus-1.5B - GGUF
- Model creator: https://huggingface.co/Ba2han/
- Original model: https://huggingface.co/Ba2han/Tinypus-1.5B/
| Name | Quant method | Size |
|---|---|---|
| Tinypus-1.5B.Q2_K.gguf | Q2_K | 0.52GB |
| Tinypus-1.5B.IQ3_XS.gguf | IQ3_XS | 0.58GB |
| Tinypus-1.5B.IQ3_S.gguf | IQ3_S | 0.61GB |
| Tinypus-1.5B.Q3_K_S.gguf | Q3_K_S | 0.61GB |
| Tinypus-1.5B.IQ3_M.gguf | IQ3_M | 0.63GB |
| Tinypus-1.5B.Q3_K.gguf | Q3_K | 0.67GB |
| Tinypus-1.5B.Q3_K_M.gguf | Q3_K_M | 0.67GB |
| Tinypus-1.5B.Q3_K_L.gguf | Q3_K_L | 0.72GB |
| Tinypus-1.5B.IQ4_XS.gguf | IQ4_XS | 0.74GB |
| Tinypus-1.5B.Q4_0.gguf | Q4_0 | 0.78GB |
| Tinypus-1.5B.IQ4_NL.gguf | IQ4_NL | 0.78GB |
| Tinypus-1.5B.Q4_K_S.gguf | Q4_K_S | 0.78GB |
| Tinypus-1.5B.Q4_K.gguf | Q4_K | 0.82GB |
| Tinypus-1.5B.Q4_K_M.gguf | Q4_K_M | 0.82GB |
| Tinypus-1.5B.Q4_1.gguf | Q4_1 | 0.86GB |
| Tinypus-1.5B.Q5_0.gguf | Q5_0 | 0.94GB |
| Tinypus-1.5B.Q5_K_S.gguf | Q5_K_S | 0.94GB |
| Tinypus-1.5B.Q5_K.gguf | Q5_K | 0.96GB |
| Tinypus-1.5B.Q5_K_M.gguf | Q5_K_M | 0.96GB |
| Tinypus-1.5B.Q5_1.gguf | Q5_1 | 1.02GB |
| Tinypus-1.5B.Q6_K.gguf | Q6_K | 1.11GB |
| Tinypus-1.5B.Q8_0.gguf | Q8_0 | 1.44GB |
Original model description:
license: mit datasets:
- garage-bAInd/Open-Platypus pipeline_tag: text-generation
*drumroll please*
Introducing Tinypus!
I passthrough merged base Tiny Llama Chat with itself, then fine-tuned with around 1/3 of Platypus dataset.
Observations:
-
It's smarter (I think?)
-
It sometimes throws "### Instruction:" line. This could be due to the platypus dataset, or the fact that I know jackshit about programming. You can add it to "custom stopping strings" in oobaboga.
-
It may be possible to train very specialized mini experts and merge them???
Template
Same with TinyLlama/TinyLlama-1.1B-Chat-v1.0
Merge details
slices:
-
sources:
-
model: E://text-generation-webui//models//TinyLlama
layer_range: [0, 12]
-
-
sources:
-
model: E://text-generation-webui//models//TinyLlama
layer_range: [4, 22]
-
merge_method: passthrough
dtype: bfloat16
QLoRA Details
Chunk Length: 1152 R/A: 64/128 Epoch: 1 q-k-v-o
- Created
- Jun 25, 2024
- Updated
- Jun 25, 2024