-
Hi,
I am finetuning tiny llama on T4 with FP16. When I use packing the loss seems to be okay. But when I set it to false, the grad_norm goes to nan and the model doesn't learn anything. This came …
-
when download model `tinyllama:gguf` the model id should be `tinyllama:gguf` instead of `tinyllama-gguf`
![Image](https://github.com/user-attachments/assets/2a1d9c0d-1c49-460b-bdce-798d9b604333)
-
Hi @songmzhang I tried running the evaluation shell script with TinyLlama and it threw the error :
Any leads into what might be the issue here? Mistral checkpoint ran fine with this modified ev…
-
-
### Description of the bug:
- ai-edge-torch version: 2.0
- command
```c
CC=/usr/bin/clang-18 bazel run -c opt //ai_edge_torch/generative/examples/c++:text_generator_main -- --tflite_model=/mnt/da…
-
1. Start the first conversation in Jan
2. Choose tinyllama-1.1b
3. It started generating nonsense.
What's wrong with tinyllama-1.1b? Is it because of the model or because of our configuration?
…
-
Whenever I give the llama3 model a prompt it takes about 5 minutes to actually respond and sometimes is just times out. When I use a smaller model like Tinyllama it goes faster but tinyllama has no id…
-
### What is the issue?
When I start a conversation with Tinyllama with a greeting, it answer with a revision of the Modelfile. That's strange. Here is an example
![chrome_screenshot_20 jul 2024 6_3…
-
-
### Your current environment
The output of `python collect_env.py`
```text
Collecting environment information...
PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch…