Text Generation
Transformers
Safetensors
Finnish
llama
finnish
conversational
text-generation-inference
File size: 2,580 Bytes
a85f909
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
947b4f4
6c26cc3
 
a85f909
947b4f4
a85f909
 
 
 
 
 
 
 
 
 
 
 
6c26cc3
 
a85f909
 
 
 
 
 
 
6c26cc3
a85f909
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
#! /bin/bash

# Put your WANDB API key here to enable logging to wandb.
export WANDB_API_KEY=''

# TPU specific flags to improve training throughput
export LIBTPU_INIT_ARGS='--xla_jf_spmd_threshold_for_windowed_einsum_mib=0 --xla_tpu_spmd_threshold_for_allgather_cse=10000 --xla_tpu_spmd_rewrite_einsum_with_reshape=true --xla_enable_async_all_gather=true --jax_enable_async_collective_offload=true --xla_tpu_enable_latency_hiding_scheduler=true TPU_MEGACORE=MEGACORE_DENSE'


python3 -m EasyLM.models.llama.llama_train \
    --jax_distributed.initialize_jax_distributed=True \
    --mesh_dim='1,-1,4' \
    --dtype='bf16' \
    --total_steps=900000 \
    --eval_freq=50000 \
    --log_freq=1000 \
    --save_model_freq=2000 \
    --save_milestone_freq=50000 \
    --load_llama_config='7b' \
    --update_llama_config='' \
    --load_dataset_state='' \
    --load_checkpoint='' \
    --tokenizer.vocab_file='tokenizer.model' \
    --optimizer.type='lion' \
    --optimizer.lion_optimizer.weight_decay=1.0 \
    --optimizer.lion_optimizer.lr_schedule_type='warmup_constant_linear_decay' \
    --optimizer.lion_optimizer.lr=1e-4 \
    --optimizer.lion_optimizer.end_lr=1e-5 \
    --optimizer.lion_optimizer.lr_warmup_steps=60000 \
    --optimizer.lion_optimizer.lr_constant_steps=900000 \
    --optimizer.lion_optimizer.lr_decay_steps=100000 \
    --optimizer.lion_optimizer.bf16_momentum=True \
    --train_dataset.type='huggingface' \
    --train_dataset.text_processor.fields='text' \
    --train_dataset.text_processor.add_eos_token=True \
    --train_dataset.text_processor.add_bos_token=True \
    --train_dataset.huggingface_dataset.path='/researchdisk/lm_training_dataset_first_stage' \
    --train_dataset.huggingface_dataset.split='train' \
    --train_dataset.huggingface_dataset.seq_length=2048 \
    --train_dataset.huggingface_dataset.batch_size=64 \
    --eval_dataset.type='huggingface' \
    --eval_dataset.text_processor.fields='text' \
    --eval_dataset.text_processor.add_eos_token=True \
    --eval_dataset.text_processor.add_bos_token=True \
    --eval_dataset.huggingface_dataset.path='/researchdisk/lm_training_dataset_first_stage' \
    --eval_dataset.huggingface_dataset.split='validation' \
    --eval_dataset.huggingface_dataset.seq_length=2048 \
    --eval_dataset.huggingface_dataset.batch_size=64 \
    --checkpointer.save_optimizer_state=True \
    --logger.online=True \
    --logger.prefix='EasyLM' \
    --logger.project="llama-7b-v2" \
    --logger.output_dir="gs://finnish-nlp-research-us/llama-7b-v2-checkpoint" \
    --logger.wandb_dir="./"