Upload README.md
Browse files
README.md
CHANGED
@@ -106,12 +106,12 @@ Most GPTQ files are made with AutoGPTQ. Mistral models are currently made with T
|
|
106 |
|
107 |
| Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
|
108 |
| ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
|
109 |
-
| [main](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/main) | 4 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/) | 2048 | 0.77 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. |
|
110 |
-
| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/) | 2048 | 0.82 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. |
|
111 |
-
| [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/) | 2048 | 1.23 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. |
|
112 |
-
| [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/) | 2048 | 1.26 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. |
|
113 |
-
| [gptq-8bit-32g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-8bit-32g-actorder_True) | 8 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/) | 2048 | 1.32 GB | No | 8-bit, with group size 32g and Act Order for maximum inference quality. |
|
114 |
-
| [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/) | 2048 | 0.79 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
|
115 |
|
116 |
<!-- README_GPTQ.md-provided-files end -->
|
117 |
|
|
|
106 |
|
107 |
| Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
|
108 |
| ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
|
109 |
+
| [main](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/main) | 4 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-raw-v1) | 2048 | 0.77 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. |
|
110 |
+
| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-raw-v1) | 2048 | 0.82 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. |
|
111 |
+
| [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-raw-v1) | 2048 | 1.23 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. |
|
112 |
+
| [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-raw-v1) | 2048 | 1.26 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. |
|
113 |
+
| [gptq-8bit-32g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-8bit-32g-actorder_True) | 8 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-raw-v1) | 2048 | 1.32 GB | No | 8-bit, with group size 32g and Act Order for maximum inference quality. |
|
114 |
+
| [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/TinyLlama-1.1B-intermediate-step-715k-1.5T-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-raw-v1) | 2048 | 0.79 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
|
115 |
|
116 |
<!-- README_GPTQ.md-provided-files end -->
|
117 |
|