LSX-UniWue/LLaMmlein_1B

LLäMmlein 1B

LLäMmlein 1B is a German LLaMa model trained from scratch using our adapted Tinyllama codebase on the German portion of RedPajama V2. To enhance data quality, we additionally deduplicated the dataset on paragraph level and filtered it using a token-to-word ratio filter. The resulting dataset can be found here.

We provide three model sizes:

• LLäMmlein 7B

• LLäMmlein 1B ← You are here

• LLäMmlein 120M

Find more details on our page our page and our preprint!

Usage

You can use LLäMmlein with the transformers library. (Optional: install flash-attn to achieve highest efficiency.)

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "LSX-UniWue/LLaMmlein_1B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Intermediate Checkpoints

In addition to the final model checkpoint, we publish intermediate checkpoints throughout the full training process as unique branches in this repository. A specific checkpoint can be loaded like this:

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "LSX-UniWue/LLaMmlein_1B"
revision = "iter-00420000-ckpt"
tokenizer = AutoTokenizer.from_pretrained(model_id, revision=revision)
model = AutoModelForCausalLM.from_pretrained(model_id, revision=revision)

Next to the model itself each branch contains all datapoints that were used to train the model up to that point. In the correspinding folder, named after the checkpoint, you can find several .log files (depending on the number of GPUs) of the following format:

{"time": 1739809392.679516, 
  "iter_num": 0, 
  "data_id": ["sha1:EDQMBYDCYBLDAZH3MGYM276BM2DEHPPJ", "sha1:SAJCI75DRHZZFGQORV66NB5FVWUAVLFH", "sha1:7RBZV2MCEM4TUGBBWGTFQAKTWUOGETZU", "sha1:234M32IMLZF7455AKOFWDP6HT6YXAYB4", "sha1:2BIZ7LLSHRK5GUGPZM2GM55APTDKBUG2", "sha1:OF7OI77ZT7ROXGMB6LL4RSRANX7REAYK", "sha1:LGPUOCOV3MKETI5F3IHVGZPD4M26NNJL", "sha1:SHIHUW7FJTP5YHFFV2JZ2CAHUVMKK7XG"], 
  "file_id": [0, 0, 0, 0, 0, 0, 0, 0], 
  "process_rank": 0}

Note: Our earlier models from the paper, which do not include data logging, are available at:

• LLäMmlein 1B prerelease

• LLäMmlein 120M prerelease

License

We release the LLäMmlein models under a research-only RAIL-M license. See license.md for details.