avtc

Results 28 comments of avtc

I have encountered similar issue, but without hitting recursion limit. Looks like it is related with ripgrep search. ![Image](https://github.com/user-attachments/assets/ced82b14-7b9e-4003-83e5-0e4b11872140)

ra-aid --chat --provider openai-compatible --model athene-v2-chat@iq3_m --temperature 0.1 Prompt: Please add Piece movement animation on UI. With ability to turn it off via constant. For human player and for AI...

The project files to reproduce issue: [AIChess-Repro.zip](https://github.com/user-attachments/files/19267963/AIChess-Repro.zip) I have reproduced the issue with qwen2.5-coder-32b as well. OS: Windows. Python 3.10.11 ripgrep installed today via: choco install ripgrep

The workaround is to ask model to not use a tool like that: Please do not use ripgrep_search tool as it is broken. For me it works after error appear,...

Encountered same issue. For me it reproduced every time with a --chat command. If I am not press the OK button in a short time - the dialog window hides...

Another workaround, allowing named constrains also for `not null bit`/`int`/etc columns without `DefaultValueSql` specified: ```CSharp public class MSqlDbContext : DbContext { protected override void OnConfiguring(DbContextOptionsBuilder optionsBuilder) { optionsBuilder .UseSqlServer("your connection...

``` INFO ModuleLooper: forward start (processor=`gptq`, layer=`model.layers.1`, subset=3/7, batches=1057) % Quantizing mlp.experts.32.gate_proj in layer [1 of 45] █------------------------------------| 0:12:39 / 4:50:57 [2/46] 4.3%Traceback (most recent call last): File "/home/ubuntu/Documents/Quantize/quantize-glm4.5-Air-gptqmodel-moe-prune-smart-4.py", line...

@Qubitium The full script. [quantize-glm4.5-Air-gptqmodel-moe-prune-smart-4.py](https://github.com/user-attachments/files/22692254/quantize-glm4.5-Air-gptqmodel-moe-prune-smart-4.py) I have lowered number of samples in: ``` c4 = load_dataset( "allenai/c4", data_files="en/c4-train.00001-of-01024.json.gz", split="train" ).shuffle(seed=42).select(range(100)) # 300 ``` and got another error, look like memory...

It there a switch to turn off data-parallelism to check without it?

After lowering samples more it passes the layer 1, will check how it goes. ``` # 1. General Language (40% - 410 samples) c4 = load_dataset( "allenai/c4", data_files="en/c4-train.00001-of-01024.json.gz", split="train" ).shuffle(seed=42).select(range(100))...