-
I am able to perform inference using the f16 model found here https://huggingface.co/nouamanetazi/bloomz-560m-ggml/tree/main but when I use `llm` to quantize it to q4_0 format, the model produces gibb…
-
Using https://huggingface.co/Wauplin/bloomz-7b1.cpp/blob/main/ggml-model-bloomz-7b1-f16-q4_0.bin:
```rust
# cargo run --release --bin llm bloom infer -m models/bloom/ggml-model-bloomz-7b1-f16-q4_0…
-
I user Macbook Pro with M2. when I run 'sh run_1.3b.sh' shows errors.
>>>
[2023-05-10 23:27:00,776] [WARNING] [runner.py:191:fetch_hostfile] Unable to find hostfile, will proceed with training with …
-
I try to put same input to different platform:
1、modelscrope:https://modelscope.cn/studios/damo/mPLUG-Owl-Bilingual/summary
2、local demo by the weight from https://huggingface.co/MAGAer13/mplug-owl-…
-
Here is the example:
``` shell
# origin example
cargo run --release --bin llm bloom infer \
--model-path ./models/ggml-model-phoenix-inst-chat-7b-f16-q4_0.bin \
--num-ctx-tokens 1024 -…
-
在sft.py 的
769行开始德循环内logger.warning(f"tokenization mismatch: {cur_len} vs. {total_len}. (ignored)") 这里
cur_len 一直为1
| WARNING | __main__:preprocess_function:813 - tokenization mismatch: 1 vs. 81.…
-
RT
-
RuntimeError: unscale_() has already been called on this optimizer since the last update().
-
Hello everyone,
I would like to inquire as to why you have chosen Bloom-7b1 as your paper description. As far as I know, BigScience recommends using Bloomz variants, which can be found by following t…
-
Does this program supports tensorboard? Could not find any logs of tensorbard.