Hugging Face's logo

latimar
/
Phind-Codellama-34B-v2-exl2

Text Generation
Transformers
llama
text-generation-inference
Model card Files Community
5
latimar commited on
Commit
68e49f9
•
1 Parent(s): 1c6adcc

Update README

Browse files
Files changed (1) hide show
  1. README.md +2 -1
README.md CHANGED
@@ -24,6 +24,7 @@ There are the following branches:
24
 
25
  ```
26
  5_0-bpw-h8
 
27
  4_625-bpw-h6
28
  4_125-bpw-h6
29
  3_8-bpw-h6
@@ -33,7 +34,7 @@ There are the following branches:
33
 
34
  * Calibration dataset used for conversion: [wikitext-v2](https://huggingface.co/datasets/wikitext/blob/refs%2Fconvert%2Fparquet/wikitext-2-v1/test/0000.parquet)
35
  * Evaluation dataset used to calculate perplexity: [wikitext-v2](https://huggingface.co/datasets/wikitext/blob/refs%2Fconvert%2Fparquet/wikitext-2-v1/validation/0000.parquet)
36
- * Calibration dataset used for conversion of `5_0-bpw-h8-ev`: [wizardLM-evol-instruct_70k](https://huggingface.co/datasets/WizardLM/WizardLM_evol_instruct_70k/blob/refs%2Fconvert%2Fparquet/default/train/0000.parquet)
37
  * Evaluation dataset used to calculate ppl for `Evol-Ins`: : [nikrosh-evol-instruct](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1/blob/refs%2Fconvert%2Fparquet/default/train/0000.parquet)
38
  * PPL max seq. length used: 1792 (2048 with 5.0-bpw-h8 causes OOM on RTX 4090 when evaluating ppl, so had to go down a bit)
39
 
 
24
 
25
  ```
26
  5_0-bpw-h8
27
+ 5_0-bpw-h8-evol-ins
28
  4_625-bpw-h6
29
  4_125-bpw-h6
30
  3_8-bpw-h6
 
34
 
35
  * Calibration dataset used for conversion: [wikitext-v2](https://huggingface.co/datasets/wikitext/blob/refs%2Fconvert%2Fparquet/wikitext-2-v1/test/0000.parquet)
36
  * Evaluation dataset used to calculate perplexity: [wikitext-v2](https://huggingface.co/datasets/wikitext/blob/refs%2Fconvert%2Fparquet/wikitext-2-v1/validation/0000.parquet)
37
+ * Calibration dataset used for conversion of `5_0-bpw-h8-evol-ins`: [wizardLM-evol-instruct_70k](https://huggingface.co/datasets/WizardLM/WizardLM_evol_instruct_70k/blob/refs%2Fconvert%2Fparquet/default/train/0000.parquet)
38
  * Evaluation dataset used to calculate ppl for `Evol-Ins`: : [nikrosh-evol-instruct](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1/blob/refs%2Fconvert%2Fparquet/default/train/0000.parquet)
39
  * PPL max seq. length used: 1792 (2048 with 5.0-bpw-h8 causes OOM on RTX 4090 when evaluating ppl, so had to go down a bit)
40