Rdoublea's workspace
Runs
76
Name
69 visualized
State
Notes
User
Tags
Created
Runtime
Sweep
base_model_path
batch_size
checkpointer._component_
checkpointer.checkpoint_dir
checkpointer.checkpoint_files
checkpointer.checkpoint_files.filename_format
checkpointer.checkpoint_files.max_filename
checkpointer.model_type
checkpointer.output_dir
clip_grad_norm
collate_fn
compile
dataset._component_
dataset.add_eos
dataset.column
dataset.packed
dataset.partition
dataset.source
dataset.split
dataset.subset
debug_logging_enabled
debug_num_samples_per_step
device
dtype
enable_activation_checkpointing
enable_activation_offloading
epochs
forward_batch_size
gamma
gradient_accumulation_steps
grpo_samples
inference.batch_size
inference.engine
inference.group_size
inference.max_generated_tokens
inference.queue_maxsize
inference.steps_before_sync
inference.temperature
inference.total_batch_size
inference.tp_size
kl_coeff
lmbda
log_every_n_steps
log_memory_stats
Crashed
Add notes...
salman-mohammadi
6m 16s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 21s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 16s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 16s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 15s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 16s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
2m 16s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
2m 15s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 1s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 15s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
46s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
8
512
2
1
1
8
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
7m 45s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
16
512
2
1
1
16
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 46s
-
/tmp/Qwen2.5-3B
-
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_async_qwen3b
-
-
-
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
true
1
-
-
-
-
-
-
-
-
-
1
vllm
16
512
8
1
1
16
1
-
-
1
-
Crashed
Add notes...
salman-mohammadi
9m 16s
-
/tmp/Qwen2.5-3B
16
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_qwen3b
1
-
false
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
-
-
cuda
bf16
true
true
10
-
-
-
16
-
-
-
-
-
-
-
-
-
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 15s
-
/tmp/Qwen2.5-3B
16
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_qwen3b
1
-
false
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
-
-
cuda
bf16
true
true
10
-
-
-
16
-
-
-
-
-
-
-
-
-
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 1s
-
/tmp/Qwen2.5-3B
16
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_qwen3b
1
-
false
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
-
-
cuda
bf16
true
true
10
-
-
-
16
-
-
-
-
-
-
-
-
-
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m 1s
-
/tmp/Qwen2.5-3B
16
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_qwen3b
1
-
false
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
-
-
cuda
bf16
true
true
10
-
-
-
16
-
-
-
-
-
-
-
-
-
-
-
1
-
Crashed
Add notes...
salman-mohammadi
1m
-
/tmp/Qwen2.5-3B
16
torchtune.training.FullModelHFCheckpointer
/tmp/Qwen2.5-3B
["model-00001-of-00002.safetensors","model-00002-of-00002.safetensors"]
-
-
QWEN2
/tmp/checkpoints/grpo_qwen3b
1
-
false
torchtune.dev.grpo.gsm8k.gsm8k_dataset
-
-
-
1-9/10
-
-
-
-
-
cuda
bf16
true
true
10
-
-
-
16
-
-
-
-
-
-
-
-
-
-
-
1
-
1-20
of 76