Salman-mohammadi's workspace
Runs
10
Name
10 visualized
State
Notes
User
Tags
Created
Runtime
Sweep
batch_size
checkpointer._component_
checkpointer.checkpoint_dir
checkpointer.checkpoint_files
checkpointer.model_type
checkpointer.output_dir
compile
dataset._component_
dataset.train_on_input
device
dtype
enable_activation_checkpointing
epochs
gradient_accumulation_steps
loss._component_
lr_scheduler._component_
lr_scheduler.num_warmup_steps
metric_logger._component_
metric_logger.log_dir
metric_logger.project
model._component_
model.apply_lora_to_mlp
model.apply_lora_to_output
model.lora_alpha
model.lora_attn_modules
model.lora_rank
optimizer._component_
optimizer.lr
optimizer.weight_decay
optimizer_in_bwd
output_dir
profiler._component_
profiler.enabled
profiler.output_dir
resume_from_checkpoint
shuffle
tokenizer._component_
tokenizer.path
log_every_n_steps
gpu_resources
loss
lr
peak_memory_active
Finished
-
salman-mohammadi
1m 16s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-7b-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/CodeLlama-7b-hf
false
torchtune.datasets.alpaca_cleaned_dataset
true
cuda
bf16
true
1
64
torch.nn.CrossEntropyLoss
torchtune.modules.get_cosine_schedule_with_warmup
100
torchtune.utils.metric_logging.WandBLogger
/tmp/qlora_code_llama2_finetune_output/torchtune_perf_tracing.json
torchtune_codellama_testing
torchtune.models.code_llama2.qlora_code_llama2_7b
false
false
16
["q_proj","v_proj"]
8
torch.optim.AdamW
0.0003
0.01
true
/tmp/qlora_code_llama2_finetune_output
torchtune.utils.profiler
false
/tmp/qlora_code_llama2_finetune_output/torchtune_perf_tracing.json
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-7b-hf/tokenizer.model
-
12647949824
1.53467
0
12.85234
Finished
-
salman-mohammadi
4s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/Llama-2-7b-hf
["pytorch_model-00001-of-00002.bin","pytorch_model-00002-of-00002.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
torchtune.datasets.alpaca_cleaned_dataset
true
cuda
bf16
true
1
16
torch.nn.CrossEntropyLoss
torchtune.modules.get_cosine_schedule_with_warmup
100
torchtune.utils.metric_logging.WandBLogger
/tmp/qlora_finetune_output/
torchtune_codellama_testing
torchtune.models.llama2.qlora_llama2_7b
true
false
16
["q_proj","v_proj","k_proj","output_proj"]
8
torch.optim.AdamW
0.0003
0.01
-
/tmp/qlora_finetune_output/
torchtune.utils.profiler
false
/tmp/qlora_finetune_output//torchtune_perf_tracing.json
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/Llama-2-7b-hf/tokenizer.model
1
-
-
-
-
Finished
-
salman-mohammadi
50s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-7b-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/CodeLlama-7b-hf
false
torchtune.datasets.alpaca_cleaned_dataset
true
cuda
bf16
true
1
64
torch.nn.CrossEntropyLoss
torchtune.modules.get_cosine_schedule_with_warmup
100
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_code_llama2_finetune_output/torchtune_perf_tracing.json
torchtune_codellama_testing
torchtune.models.code_llama2.lora_code_llama2_7b
false
false
16
["q_proj","v_proj"]
8
torch.optim.AdamW
0.0003
0.01
true
/tmp/lora_code_llama2_finetune_output
torchtune.utils.profiler
false
/tmp/lora_code_llama2_finetune_output/torchtune_perf_tracing.json
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-7b-hf/tokenizer.model
-
14211938816
1.62569
0
14.13296
Finished
-
salman-mohammadi
52s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-7b-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/CodeLlama-7b-hf
false
torchtune.datasets.alpaca_cleaned_dataset
true
cuda
bf16
true
1
64
torch.nn.CrossEntropyLoss
torchtune.modules.get_cosine_schedule_with_warmup
100
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_code_llama2_finetune_output/torchtune_perf_tracing.json
torchtune_codellama_testing
torchtune.models.code_llama2.lora_code_llama2_7b
false
false
16
["q_proj","v_proj"]
8
torch.optim.AdamW
0.0003
0.01
true
/tmp/lora_code_llama2_finetune_output
-
-
-
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-7b-hf/tokenizer.model
-
-
-
-
-
Finished
-
salman-mohammadi
15s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-7b-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/CodeLlama-7b-hf
false
torchtune.datasets.alpaca_cleaned_dataset
true
cuda
bf16
true
1
64
torch.nn.CrossEntropyLoss
torchtune.modules.get_cosine_schedule_with_warmup
100
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_code_llama2_finetune_output/torchtune_perf_tracing.json
torchtune_codellama_testing
torchtune.models.lora_code_llama2.code_llama2_7b
false
false
16
["q_proj","v_proj"]
8
torch.optim.AdamW
0.0003
0.01
true
/tmp/lora_code_llama2_finetune_output
-
-
-
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-7b-hf/tokenizer.model
-
-
-
-
-
Finished
-
salman-mohammadi
2m 32s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-7b-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/CodeLlama-7b-hf
false
torchtune.datasets.alpaca_dataset
true
cuda
bf16
true
3
1
torch.nn.CrossEntropyLoss
-
-
torchtune.utils.metric_logging.WandBLogger
/tmp/code_llama2_finetune
torchtune_codellama_testing
torchtune.models.code_llama2.code_llama2_7b
-
-
-
-
-
bitsandbytes.optim.PagedAdamW
0.00002
-
true
/tmp/code_llama2_finetune
-
-
-
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-7b-hf/tokenizer.model
-
13902832640
1.24063
0.00002
-
Finished
-
salman-mohammadi
4s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-7b-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/CodeLlama-7b-hf
false
torchtune.datasets.alpaca_dataset
true
cuda
bf16
true
3
1
torch.nn.CrossEntropyLoss
-
-
torchtune.utils.metric_logging.WandBLogger
/tmp/code_llama2_finetune
torchtune_codellama_testing
torchtune.models.code_llama2.code_llama2_7b
-
-
-
-
-
bitsandbytes.optim.PagedAdamW
0.00002
-
true
/tmp/code_llama2_finetune
-
-
-
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-7b-hf/tokenizer.model
-
-
-
-
-
Finished
-
salman-mohammadi
41m 41s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-7b-Instruct-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
torchtune.datasets.alpaca_cleaned_dataset
true
cuda
bf16
true
1
64
torch.nn.CrossEntropyLoss
torchtune.modules.get_cosine_schedule_with_warmup
100
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_finetune_output
torchtune_codellama_testing
torchtune.models.llama2.lora_code_llama2_7b
false
false
16
["q_proj","v_proj"]
8
torch.optim.AdamW
0.0003
0.01
-
/tmp/lora_finetune_output
torchtune.utils.profiler
false
/tmp/lora_finetune_output/torchtune_perf_tracing.json
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-7b-Instruct-hf/tokenizer.model
-
13919330304
1.41144
0.000165
14.58847
Finished
-
salman-mohammadi
58m 11s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-13b-Instruct-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/Llama-2-13b-hf/
false
torchtune.datasets.alpaca_cleaned_dataset
true
cuda
bf16
true
1
16
torch.nn.CrossEntropyLoss
torchtune.modules.get_cosine_schedule_with_warmup
100
torchtune.utils.metric_logging.WandBLogger
/tmp/qlora_finetune_output/
torchtune_codellama_testing
torchtune.models.llama2.qlora_code_llama2_13b
true
false
16
["q_proj","v_proj","k_proj","output_proj"]
8
torch.optim.AdamW
0.0003
0.01
-
/tmp/qlora_finetune_output/
torchtune.utils.profiler
false
/tmp/qlora_finetune_output//torchtune_perf_tracing.json
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-13b-Instruct-hf/tokenizer.model
1
8594478080
1.29708
0.000093
9.45774
Finished
-
salman-mohammadi
24m 4s
-
2
torchtune.utils.FullModelHFCheckpointer
/tmp/CodeLlama-7b-Instruct-hf
["pytorch_model-00001-of-00003.bin","pytorch_model-00002-of-00003.bin","pytorch_model-00003-of-00003.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
torchtune.datasets.alpaca_dataset
true
cuda
bf16
true
3
1
torch.nn.CrossEntropyLoss
-
-
torchtune.utils.metric_logging.WandBLogger
/tmp/alpaca-llama2-finetune
torchtune_codellama_testing
torchtune.models.llama2.code_llama2_7b
-
-
-
-
-
bitsandbytes.optim.PagedAdamW
0.00002
-
true
/tmp/alpaca-llama2-finetune
-
-
-
false
true
torchtune.models.llama2.llama2_tokenizer
/tmp/CodeLlama-7b-Instruct-hf/tokenizer.model
-
14020930048
1.31372
0.00002
14.44884
1-10
of 10