Salman-mohammadi's workspace
Runs
97
Name
2 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
compile_model
dataset._component_
dataset._max_seq_len
dataset.chat_format
dataset.conversation_style
dataset.max_seq_len
dataset.source
dataset.split
device
dtype
enable_activation_checkpointing
epochs
gradient_accumulation_steps
log_every_n_steps
log_memory_stats
log_peak_memory_stats
loss._component_
loss.beta
loss.gamma
loss.label_smoothing
lr_scheduler._component_
lr_scheduler.num_warmup_steps
max_steps_per_epoch
metric_logger._component_
metric_logger.log_dir
metric_logger.project
model._component_
model.apply_lora_to_mlp
model.apply_lora_to_output
model.attn_dropout
model.embed_dim
model.intermediate_dim
model.lora_alpha
model.lora_attn_modules
model.lora_dropout
model.lora_rank
model.max_seq_len
Killed
-
salman-mohammadi
2m 9s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
1m 40s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
11m 59s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
4m 51s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
3
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
2m 54s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
3
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
9m 17s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
2.5
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
1m 41s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
2.5
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
42m 25s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
3m 51s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
16
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
3m 36s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
16
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
4m 30s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
1
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
17m 36s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
3m 27s
-
8
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
16
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
200
2000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
2m 2s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
140
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
400
4000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
33m 59s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
40
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
400
4000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Failed
-
salman-mohammadi
7m 34s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
40
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
400
4000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Failed
-
salman-mohammadi
1m 33s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
40
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
400
4000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Finished
-
salman-mohammadi
9s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
40
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
400
4000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
1m 4s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
40
8
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
400
4000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
Killed
-
salman-mohammadi
15s
-
16
torchtune.utils.FullModelHFCheckpointer
./target/dummy
["pytorch_model.bin"]
LLAMA2
/tmp/Llama-2-7b-hf
false
-
torchtune.datasets.stack_exchanged_paired_dataset
-
-
-
256
-
-
mps
bf16
true
40
16
1
-
false
torchtune.modules.loss.SimPOLoss
2
0.5
0
torchtune.modules.get_cosine_schedule_with_warmup
400
4000
torchtune.utils.metric_logging.WandBLogger
/tmp/lora_dpo_output/
torchtune_simpo
torchtune.models.llama2.lora_llama2
true
false
0
2048
5632
16
["q_proj","v_proj","k_proj","output_proj"]
0
32
2048
1-20
of 97