llm-foundry

llm-foundry copied to clipboard

Generated sample equals to input samples

3

I have trained 125m MPT on some small dataset, my generated inputs via `inference/hf_generate.py` (before this converted from composer to HF, and it's gives me some value from `eval/eval.py` with...

germanjke

unexpected results in inference

5

i am trying to inference mosaicml/mpt-7b model on colab but got unexpected results ``` !python /content/llm-foundry/scripts/inference/hf_generate.py \ --name_or_path 'mosaicml/mpt-7b' \ --temperature 1.0 \ --top_p 0.95 \ --top_k 50 \ --seed...

OmarMohammed88

keyError : 'attn_impl' when converting a composer checkpoint to a standard HF checkpoint

3

I uncomment the following lines in https://github.com/mosaicml/llm-foundry/blob/main/scripts/train/finetune_example/gpt2-arc-easy.yaml save_num_checkpoints_to_keep: 1 # Important, this cleans up checkpoints saved to DISK save_folder: ./{run_name}/checkpoints Then run "composer train.py finetune_example/gpt2-arc-easy.yaml". I can see the saved...

jdongca2003

Triton Test Failed: GPU SMs must run at 1350 MHz / GPU memory must run at 877 MHz

3

Hi, I am trying to run the tests suite to see if my setup is correct and I am down to 31 failed, 4852 passed etc... However the ones that...

cocobeach

Multi-nodes slurm training

1

I wanna ask about training with slum command. I'm training 7b parameters model but apparently when i set the environment with more than one node it does see only one...

j-Gaow

Changes almost identical to those in https://github.com/mosaicml/examples/pull/335

aspfohl

Why does convert_dataset_json.py only support 'train' for the --split argument?"

1


sysusicily

Not an issue, just a question. Does llm-foundry automatically handle eos tokens or should we manually add them into our text data to denote? For example, if we are loading...

tginart

downloading datasets

1

Hi! you have script for prepare data in your `scripts/train` which is `python ../data_prep/convert_dataset_hf.py --dataset c4 --data_subset en --out_root ./my-copy-c4 --splits train_small val_small --concat_tokens 2048 --tokenizer EleutherAI/gpt-neox-20b --eos_text ''` can...

germanjke

Upgrade to `mosaicml-streaming==0.5.x`

2

abhi-mosaic