Delete 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1
Browse files- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/.hydra/config.yaml +0 -115
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/.hydra/hydra.yaml +0 -154
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/.hydra/overrides.yaml +0 -1
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/config.json +0 -42
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/generation_config.json +0 -7
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/merges.txt +0 -0
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/mlp_projector.bin +0 -3
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/model.safetensors +0 -3
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/optimizer.pt +0 -3
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/rng_state.pth +0 -3
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/scheduler.pt +0 -3
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/special_tokens_map.json +0 -34
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/tokenizer.json +0 -0
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/tokenizer_config.json +0 -155
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/trainer_state.json +0 -2841
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/training_args.bin +0 -3
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/vocab.json +0 -0
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/config.json +0 -42
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/generation_config.json +0 -7
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/model.safetensors +0 -3
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/normalizer.pt +0 -3
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/train.log +0 -11
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/debug-internal.log +0 -17
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/debug.log +0 -35
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/files/config.yaml +0 -711
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/files/output.log +0 -509
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/files/wandb-metadata.json +0 -55
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/files/wandb-summary.json +0 -1
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug-core.log +0 -16
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug-internal.log +0 -17
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug.log +0 -35
- 2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/run-nhmfpc2t.wandb +0 -3
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/.hydra/config.yaml
DELETED
|
@@ -1,115 +0,0 @@
|
|
| 1 |
-
name: train_llm_lowdim
|
| 2 |
-
_target_: llmbc.workspace.train_llm_workspace.TrainLLMWorkspace
|
| 3 |
-
obs_dim: ${task.obs_dim}
|
| 4 |
-
action_dim: ${task.action_dim}
|
| 5 |
-
horizon: 1
|
| 6 |
-
n_obs_steps: 1
|
| 7 |
-
n_action_steps: 1
|
| 8 |
-
task_name: ${task.name}
|
| 9 |
-
exp_name: train llm
|
| 10 |
-
model_name: ${llm.name}
|
| 11 |
-
use_quantization: ${llm.use_quantization}
|
| 12 |
-
lora_config: ${llm.lora_config}
|
| 13 |
-
dataset:
|
| 14 |
-
test_data_ratio: 0.01
|
| 15 |
-
debug: false
|
| 16 |
-
training:
|
| 17 |
-
seed: 42
|
| 18 |
-
per_device_train_batch_size: 128
|
| 19 |
-
per_device_eval_batch_size: 128
|
| 20 |
-
gradient_accumulation_steps: 1
|
| 21 |
-
optim: paged_adamw_32bit
|
| 22 |
-
num_train_epochs: 10
|
| 23 |
-
eval_strategy: steps
|
| 24 |
-
logging_steps: 1
|
| 25 |
-
warmup_steps: 10
|
| 26 |
-
logging_strategy: steps
|
| 27 |
-
learning_rate: 0.0001
|
| 28 |
-
fp16: false
|
| 29 |
-
bf16: true
|
| 30 |
-
tf32: true
|
| 31 |
-
group_by_length: true
|
| 32 |
-
report_to: wandb
|
| 33 |
-
save_steps: 5000
|
| 34 |
-
eval_steps: 10
|
| 35 |
-
use_joint_mlp_projector: ${llm.use_joint_mlp_projector}
|
| 36 |
-
joint_obs_action_mlp_lr: 5.0e-05
|
| 37 |
-
trainer:
|
| 38 |
-
obs_dim: ${obs_dim}
|
| 39 |
-
action_dim: ${action_dim}
|
| 40 |
-
use_joint_mlp_projector: ${llm.use_joint_mlp_projector}
|
| 41 |
-
max_seq_length: ${llm.max_length}
|
| 42 |
-
dataset_text_field: text
|
| 43 |
-
packing: false
|
| 44 |
-
logging:
|
| 45 |
-
project: llm_module_finetuning
|
| 46 |
-
resume: true
|
| 47 |
-
mode: online
|
| 48 |
-
name: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
|
| 49 |
-
tags:
|
| 50 |
-
- ${name}
|
| 51 |
-
- ${task_name}
|
| 52 |
-
- ${exp_name}
|
| 53 |
-
id: null
|
| 54 |
-
group: null
|
| 55 |
-
multi_run:
|
| 56 |
-
run_dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
|
| 57 |
-
wandb_name_base: ${now:%Y.%m.%d-%H.%M.%S}_${name}_${task_name}
|
| 58 |
-
task:
|
| 59 |
-
name: adroit-hand-hammer-v1
|
| 60 |
-
obs_dim: 46
|
| 61 |
-
action_dim: 26
|
| 62 |
-
env_runner:
|
| 63 |
-
_target_: llmbc.env_runner.adroit_lowdim_runner.AdroitHandLowdimRunner
|
| 64 |
-
env_name: llf-adroit-adroit-hand-hammer-v1
|
| 65 |
-
n_train: 10
|
| 66 |
-
n_test: 50
|
| 67 |
-
n_envs: 10
|
| 68 |
-
max_steps: 150
|
| 69 |
-
n_obs_steps: ${n_obs_steps}
|
| 70 |
-
n_action_steps: ${n_action_steps}
|
| 71 |
-
instruction_type: b
|
| 72 |
-
feedback_type:
|
| 73 |
-
- hp
|
| 74 |
-
- hn
|
| 75 |
-
- fp
|
| 76 |
-
visual: false
|
| 77 |
-
discount: 0.99
|
| 78 |
-
dataset:
|
| 79 |
-
_target_: llmbc.dataset.adroit_lowdim_dataset.AdroitHandLowdimDataset
|
| 80 |
-
data_path: datasets/adroit-hand-hammer-v1-general.pt
|
| 81 |
-
data_path2: datasets/adroit-hand-hammer-v1.pt
|
| 82 |
-
horizon: ${horizon}
|
| 83 |
-
pad_before: ${eval:'${n_obs_steps}-1'}
|
| 84 |
-
pad_after: ${eval:'${n_action_steps}-1'}
|
| 85 |
-
obs_eef_target: true
|
| 86 |
-
use_manual_normalizer: false
|
| 87 |
-
val_ratio: 0.05
|
| 88 |
-
dummy_normalizer: false
|
| 89 |
-
instructor:
|
| 90 |
-
_target_: llmbc.translator.instructor.adroit_instructor.adroit_hand_hammer_v1_instructor.AdroitHandHammerV1Instructor
|
| 91 |
-
llm:
|
| 92 |
-
name: HuggingFaceTB/SmolLM2-135M-Instruct
|
| 93 |
-
model_name: SmolLM2-135M-Instruct
|
| 94 |
-
config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
|
| 95 |
-
causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
|
| 96 |
-
use_quantization: false
|
| 97 |
-
use_joint_mlp_projector: true
|
| 98 |
-
llm_mode: mlp-finetuned
|
| 99 |
-
finetune_mode: orig
|
| 100 |
-
checkpoint: data/outputs/2026.03.27/14.38.20_train_mlp_projector_adroit-hand-hammer-v1/checkpoints/latest.ckpt
|
| 101 |
-
max_length: 100
|
| 102 |
-
lora_config:
|
| 103 |
-
r: 32
|
| 104 |
-
lora_alpha: 64
|
| 105 |
-
lora_dropout: 0.05
|
| 106 |
-
bias: none
|
| 107 |
-
task_type: CAUSAL_LM
|
| 108 |
-
prompter:
|
| 109 |
-
_target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
|
| 110 |
-
use_joint_mlp_projector: true
|
| 111 |
-
hydra:
|
| 112 |
-
job:
|
| 113 |
-
override_dirname: ${model_name}
|
| 114 |
-
run:
|
| 115 |
-
dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${model_name}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/.hydra/hydra.yaml
DELETED
|
@@ -1,154 +0,0 @@
|
|
| 1 |
-
hydra:
|
| 2 |
-
run:
|
| 3 |
-
dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
|
| 4 |
-
sweep:
|
| 5 |
-
dir: data/outputs/${now:%Y.%m.%d}/${now:%H.%M.%S}_${name}_${task_name}
|
| 6 |
-
subdir: ${hydra.job.num}
|
| 7 |
-
launcher:
|
| 8 |
-
_target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
|
| 9 |
-
sweeper:
|
| 10 |
-
_target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
|
| 11 |
-
max_batch_size: null
|
| 12 |
-
params: null
|
| 13 |
-
help:
|
| 14 |
-
app_name: ${hydra.job.name}
|
| 15 |
-
header: '${hydra.help.app_name} is powered by Hydra.
|
| 16 |
-
|
| 17 |
-
'
|
| 18 |
-
footer: 'Powered by Hydra (https://hydra.cc)
|
| 19 |
-
|
| 20 |
-
Use --hydra-help to view Hydra specific help
|
| 21 |
-
|
| 22 |
-
'
|
| 23 |
-
template: '${hydra.help.header}
|
| 24 |
-
|
| 25 |
-
== Configuration groups ==
|
| 26 |
-
|
| 27 |
-
Compose your configuration from those groups (group=option)
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
$APP_CONFIG_GROUPS
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
== Config ==
|
| 34 |
-
|
| 35 |
-
Override anything in the config (foo.bar=value)
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
$CONFIG
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
${hydra.help.footer}
|
| 42 |
-
|
| 43 |
-
'
|
| 44 |
-
hydra_help:
|
| 45 |
-
template: 'Hydra (${hydra.runtime.version})
|
| 46 |
-
|
| 47 |
-
See https://hydra.cc for more info.
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
== Flags ==
|
| 51 |
-
|
| 52 |
-
$FLAGS_HELP
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
== Configuration groups ==
|
| 56 |
-
|
| 57 |
-
Compose your configuration from those groups (For example, append hydra/job_logging=disabled
|
| 58 |
-
to command line)
|
| 59 |
-
|
| 60 |
-
|
| 61 |
-
$HYDRA_CONFIG_GROUPS
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
Use ''--cfg hydra'' to Show the Hydra config.
|
| 65 |
-
|
| 66 |
-
'
|
| 67 |
-
hydra_help: ???
|
| 68 |
-
hydra_logging:
|
| 69 |
-
version: 1
|
| 70 |
-
formatters:
|
| 71 |
-
simple:
|
| 72 |
-
format: '[%(asctime)s][HYDRA] %(message)s'
|
| 73 |
-
handlers:
|
| 74 |
-
console:
|
| 75 |
-
class: logging.StreamHandler
|
| 76 |
-
formatter: simple
|
| 77 |
-
stream: ext://sys.stdout
|
| 78 |
-
root:
|
| 79 |
-
level: INFO
|
| 80 |
-
handlers:
|
| 81 |
-
- console
|
| 82 |
-
loggers:
|
| 83 |
-
logging_example:
|
| 84 |
-
level: DEBUG
|
| 85 |
-
disable_existing_loggers: false
|
| 86 |
-
job_logging:
|
| 87 |
-
version: 1
|
| 88 |
-
formatters:
|
| 89 |
-
simple:
|
| 90 |
-
format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
|
| 91 |
-
handlers:
|
| 92 |
-
console:
|
| 93 |
-
class: logging.StreamHandler
|
| 94 |
-
formatter: simple
|
| 95 |
-
stream: ext://sys.stdout
|
| 96 |
-
file:
|
| 97 |
-
class: logging.FileHandler
|
| 98 |
-
formatter: simple
|
| 99 |
-
filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
|
| 100 |
-
root:
|
| 101 |
-
level: INFO
|
| 102 |
-
handlers:
|
| 103 |
-
- console
|
| 104 |
-
- file
|
| 105 |
-
disable_existing_loggers: false
|
| 106 |
-
env: {}
|
| 107 |
-
mode: RUN
|
| 108 |
-
searchpath: []
|
| 109 |
-
callbacks: {}
|
| 110 |
-
output_subdir: .hydra
|
| 111 |
-
overrides:
|
| 112 |
-
hydra:
|
| 113 |
-
- hydra.mode=RUN
|
| 114 |
-
task: []
|
| 115 |
-
job:
|
| 116 |
-
name: train
|
| 117 |
-
chdir: null
|
| 118 |
-
override_dirname: ''
|
| 119 |
-
id: ???
|
| 120 |
-
num: ???
|
| 121 |
-
config_name: llmdp_llm_adroit-hand-hammer-v1.yaml
|
| 122 |
-
env_set: {}
|
| 123 |
-
env_copy: []
|
| 124 |
-
config:
|
| 125 |
-
override_dirname:
|
| 126 |
-
kv_sep: '='
|
| 127 |
-
item_sep: ','
|
| 128 |
-
exclude_keys: []
|
| 129 |
-
runtime:
|
| 130 |
-
version: 1.2.0
|
| 131 |
-
version_base: '1.2'
|
| 132 |
-
cwd: /tmp2/chyang/workspace/LLM-BC
|
| 133 |
-
config_sources:
|
| 134 |
-
- path: hydra.conf
|
| 135 |
-
schema: pkg
|
| 136 |
-
provider: hydra
|
| 137 |
-
- path: /tmp2/chyang/workspace/LLM-BC/config/main_table
|
| 138 |
-
schema: file
|
| 139 |
-
provider: main
|
| 140 |
-
- path: ''
|
| 141 |
-
schema: structured
|
| 142 |
-
provider: schema
|
| 143 |
-
output_dir: /tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1
|
| 144 |
-
choices:
|
| 145 |
-
hydra/env: default
|
| 146 |
-
hydra/callbacks: null
|
| 147 |
-
hydra/job_logging: default
|
| 148 |
-
hydra/hydra_logging: default
|
| 149 |
-
hydra/hydra_help: default
|
| 150 |
-
hydra/help: default
|
| 151 |
-
hydra/sweeper: basic
|
| 152 |
-
hydra/launcher: basic
|
| 153 |
-
hydra/output: default
|
| 154 |
-
verbose: false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/.hydra/overrides.yaml
DELETED
|
@@ -1 +0,0 @@
|
|
| 1 |
-
[]
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/config.json
DELETED
|
@@ -1,42 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_name_or_path": "HuggingFaceTB/SmolLM2-135M-Instruct",
|
| 3 |
-
"action_dim": 26,
|
| 4 |
-
"architectures": [
|
| 5 |
-
"LowdimLlamaForCausalLM"
|
| 6 |
-
],
|
| 7 |
-
"attention_bias": false,
|
| 8 |
-
"attention_dropout": 0.0,
|
| 9 |
-
"bos_token_id": 1,
|
| 10 |
-
"eos_token_id": 2,
|
| 11 |
-
"head_dim": 64,
|
| 12 |
-
"hidden_act": "silu",
|
| 13 |
-
"hidden_size": 576,
|
| 14 |
-
"initializer_range": 0.041666666666666664,
|
| 15 |
-
"intermediate_size": 1536,
|
| 16 |
-
"is_llama_config": true,
|
| 17 |
-
"max_position_embeddings": 8192,
|
| 18 |
-
"mlp_bias": false,
|
| 19 |
-
"model_type": "llama_lowdim",
|
| 20 |
-
"num_attention_heads": 9,
|
| 21 |
-
"num_hidden_layers": 30,
|
| 22 |
-
"num_key_value_heads": 3,
|
| 23 |
-
"obs_dim": 46,
|
| 24 |
-
"pad_token_id": 2,
|
| 25 |
-
"pretraining_tp": 1,
|
| 26 |
-
"rms_norm_eps": 1e-05,
|
| 27 |
-
"rope_interleaved": false,
|
| 28 |
-
"rope_scaling": null,
|
| 29 |
-
"rope_theta": 100000,
|
| 30 |
-
"tie_word_embeddings": true,
|
| 31 |
-
"torch_dtype": "float32",
|
| 32 |
-
"transformers.js_config": {
|
| 33 |
-
"kv_cache_dtype": {
|
| 34 |
-
"fp16": "float16",
|
| 35 |
-
"q4f16": "float16"
|
| 36 |
-
}
|
| 37 |
-
},
|
| 38 |
-
"transformers_version": "4.47.1",
|
| 39 |
-
"use_cache": false,
|
| 40 |
-
"use_joint_mlp_projector": true,
|
| 41 |
-
"vocab_size": 49152
|
| 42 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/generation_config.json
DELETED
|
@@ -1,7 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_from_model_config": true,
|
| 3 |
-
"bos_token_id": 1,
|
| 4 |
-
"eos_token_id": 2,
|
| 5 |
-
"pad_token_id": 2,
|
| 6 |
-
"transformers_version": "4.47.1"
|
| 7 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/merges.txt
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/mlp_projector.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:c4338608ed11f67ca4d076f1596453801d07110a1ba10d38b55690336de76e74
|
| 3 |
-
size 1499776
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/model.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:532f9325616dbe87be8846d0b03f80c41a2de5d5a05c7397eb15b3484410ede9
|
| 3 |
-
size 539588496
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/optimizer.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:7bc7e722db0999c1636e94a830207f155cf22d2c9d28f129bc987afa57af7a45
|
| 3 |
-
size 1079284794
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/rng_state.pth
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:a8c41ff260c47496f4f1ca68f9e267daacc3461dc3a79663c2d83c1ae8fbf495
|
| 3 |
-
size 14244
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/scheduler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:36cd9db9ff66ee06ad412fe768ecde23049517e84a863a3dfc041789e2c58298
|
| 3 |
-
size 1064
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/special_tokens_map.json
DELETED
|
@@ -1,34 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"additional_special_tokens": [
|
| 3 |
-
"<|im_start|>",
|
| 4 |
-
"<|im_end|>"
|
| 5 |
-
],
|
| 6 |
-
"bos_token": {
|
| 7 |
-
"content": "<|im_start|>",
|
| 8 |
-
"lstrip": false,
|
| 9 |
-
"normalized": false,
|
| 10 |
-
"rstrip": false,
|
| 11 |
-
"single_word": false
|
| 12 |
-
},
|
| 13 |
-
"eos_token": {
|
| 14 |
-
"content": "<|im_end|>",
|
| 15 |
-
"lstrip": false,
|
| 16 |
-
"normalized": false,
|
| 17 |
-
"rstrip": false,
|
| 18 |
-
"single_word": false
|
| 19 |
-
},
|
| 20 |
-
"pad_token": {
|
| 21 |
-
"content": "<|im_end|>",
|
| 22 |
-
"lstrip": false,
|
| 23 |
-
"normalized": false,
|
| 24 |
-
"rstrip": false,
|
| 25 |
-
"single_word": false
|
| 26 |
-
},
|
| 27 |
-
"unk_token": {
|
| 28 |
-
"content": "<|endoftext|>",
|
| 29 |
-
"lstrip": false,
|
| 30 |
-
"normalized": false,
|
| 31 |
-
"rstrip": false,
|
| 32 |
-
"single_word": false
|
| 33 |
-
}
|
| 34 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/tokenizer.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/tokenizer_config.json
DELETED
|
@@ -1,155 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"add_prefix_space": false,
|
| 3 |
-
"added_tokens_decoder": {
|
| 4 |
-
"0": {
|
| 5 |
-
"content": "<|endoftext|>",
|
| 6 |
-
"lstrip": false,
|
| 7 |
-
"normalized": false,
|
| 8 |
-
"rstrip": false,
|
| 9 |
-
"single_word": false,
|
| 10 |
-
"special": true
|
| 11 |
-
},
|
| 12 |
-
"1": {
|
| 13 |
-
"content": "<|im_start|>",
|
| 14 |
-
"lstrip": false,
|
| 15 |
-
"normalized": false,
|
| 16 |
-
"rstrip": false,
|
| 17 |
-
"single_word": false,
|
| 18 |
-
"special": true
|
| 19 |
-
},
|
| 20 |
-
"2": {
|
| 21 |
-
"content": "<|im_end|>",
|
| 22 |
-
"lstrip": false,
|
| 23 |
-
"normalized": false,
|
| 24 |
-
"rstrip": false,
|
| 25 |
-
"single_word": false,
|
| 26 |
-
"special": true
|
| 27 |
-
},
|
| 28 |
-
"3": {
|
| 29 |
-
"content": "<repo_name>",
|
| 30 |
-
"lstrip": false,
|
| 31 |
-
"normalized": false,
|
| 32 |
-
"rstrip": false,
|
| 33 |
-
"single_word": false,
|
| 34 |
-
"special": true
|
| 35 |
-
},
|
| 36 |
-
"4": {
|
| 37 |
-
"content": "<reponame>",
|
| 38 |
-
"lstrip": false,
|
| 39 |
-
"normalized": false,
|
| 40 |
-
"rstrip": false,
|
| 41 |
-
"single_word": false,
|
| 42 |
-
"special": true
|
| 43 |
-
},
|
| 44 |
-
"5": {
|
| 45 |
-
"content": "<file_sep>",
|
| 46 |
-
"lstrip": false,
|
| 47 |
-
"normalized": false,
|
| 48 |
-
"rstrip": false,
|
| 49 |
-
"single_word": false,
|
| 50 |
-
"special": true
|
| 51 |
-
},
|
| 52 |
-
"6": {
|
| 53 |
-
"content": "<filename>",
|
| 54 |
-
"lstrip": false,
|
| 55 |
-
"normalized": false,
|
| 56 |
-
"rstrip": false,
|
| 57 |
-
"single_word": false,
|
| 58 |
-
"special": true
|
| 59 |
-
},
|
| 60 |
-
"7": {
|
| 61 |
-
"content": "<gh_stars>",
|
| 62 |
-
"lstrip": false,
|
| 63 |
-
"normalized": false,
|
| 64 |
-
"rstrip": false,
|
| 65 |
-
"single_word": false,
|
| 66 |
-
"special": true
|
| 67 |
-
},
|
| 68 |
-
"8": {
|
| 69 |
-
"content": "<issue_start>",
|
| 70 |
-
"lstrip": false,
|
| 71 |
-
"normalized": false,
|
| 72 |
-
"rstrip": false,
|
| 73 |
-
"single_word": false,
|
| 74 |
-
"special": true
|
| 75 |
-
},
|
| 76 |
-
"9": {
|
| 77 |
-
"content": "<issue_comment>",
|
| 78 |
-
"lstrip": false,
|
| 79 |
-
"normalized": false,
|
| 80 |
-
"rstrip": false,
|
| 81 |
-
"single_word": false,
|
| 82 |
-
"special": true
|
| 83 |
-
},
|
| 84 |
-
"10": {
|
| 85 |
-
"content": "<issue_closed>",
|
| 86 |
-
"lstrip": false,
|
| 87 |
-
"normalized": false,
|
| 88 |
-
"rstrip": false,
|
| 89 |
-
"single_word": false,
|
| 90 |
-
"special": true
|
| 91 |
-
},
|
| 92 |
-
"11": {
|
| 93 |
-
"content": "<jupyter_start>",
|
| 94 |
-
"lstrip": false,
|
| 95 |
-
"normalized": false,
|
| 96 |
-
"rstrip": false,
|
| 97 |
-
"single_word": false,
|
| 98 |
-
"special": true
|
| 99 |
-
},
|
| 100 |
-
"12": {
|
| 101 |
-
"content": "<jupyter_text>",
|
| 102 |
-
"lstrip": false,
|
| 103 |
-
"normalized": false,
|
| 104 |
-
"rstrip": false,
|
| 105 |
-
"single_word": false,
|
| 106 |
-
"special": true
|
| 107 |
-
},
|
| 108 |
-
"13": {
|
| 109 |
-
"content": "<jupyter_code>",
|
| 110 |
-
"lstrip": false,
|
| 111 |
-
"normalized": false,
|
| 112 |
-
"rstrip": false,
|
| 113 |
-
"single_word": false,
|
| 114 |
-
"special": true
|
| 115 |
-
},
|
| 116 |
-
"14": {
|
| 117 |
-
"content": "<jupyter_output>",
|
| 118 |
-
"lstrip": false,
|
| 119 |
-
"normalized": false,
|
| 120 |
-
"rstrip": false,
|
| 121 |
-
"single_word": false,
|
| 122 |
-
"special": true
|
| 123 |
-
},
|
| 124 |
-
"15": {
|
| 125 |
-
"content": "<jupyter_script>",
|
| 126 |
-
"lstrip": false,
|
| 127 |
-
"normalized": false,
|
| 128 |
-
"rstrip": false,
|
| 129 |
-
"single_word": false,
|
| 130 |
-
"special": true
|
| 131 |
-
},
|
| 132 |
-
"16": {
|
| 133 |
-
"content": "<empty_output>",
|
| 134 |
-
"lstrip": false,
|
| 135 |
-
"normalized": false,
|
| 136 |
-
"rstrip": false,
|
| 137 |
-
"single_word": false,
|
| 138 |
-
"special": true
|
| 139 |
-
}
|
| 140 |
-
},
|
| 141 |
-
"additional_special_tokens": [
|
| 142 |
-
"<|im_start|>",
|
| 143 |
-
"<|im_end|>"
|
| 144 |
-
],
|
| 145 |
-
"bos_token": "<|im_start|>",
|
| 146 |
-
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 147 |
-
"clean_up_tokenization_spaces": false,
|
| 148 |
-
"eos_token": "<|im_end|>",
|
| 149 |
-
"extra_special_tokens": {},
|
| 150 |
-
"model_max_length": 8192,
|
| 151 |
-
"pad_token": "<|im_end|>",
|
| 152 |
-
"tokenizer_class": "GPT2Tokenizer",
|
| 153 |
-
"unk_token": "<|endoftext|>",
|
| 154 |
-
"vocab_size": 49152
|
| 155 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/trainer_state.json
DELETED
|
@@ -1,2841 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": null,
|
| 3 |
-
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 10.0,
|
| 5 |
-
"eval_steps": 10,
|
| 6 |
-
"global_step": 360,
|
| 7 |
-
"is_hyper_param_search": false,
|
| 8 |
-
"is_local_process_zero": true,
|
| 9 |
-
"is_world_process_zero": true,
|
| 10 |
-
"log_history": [
|
| 11 |
-
{
|
| 12 |
-
"epoch": 0.027777777777777776,
|
| 13 |
-
"grad_norm": 13.871646881103516,
|
| 14 |
-
"learning_rate": 1e-05,
|
| 15 |
-
"loss": 1.6431,
|
| 16 |
-
"step": 1
|
| 17 |
-
},
|
| 18 |
-
{
|
| 19 |
-
"epoch": 0.05555555555555555,
|
| 20 |
-
"grad_norm": 17.890308380126953,
|
| 21 |
-
"learning_rate": 2e-05,
|
| 22 |
-
"loss": 1.6018,
|
| 23 |
-
"step": 2
|
| 24 |
-
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 0.08333333333333333,
|
| 27 |
-
"grad_norm": 13.746294021606445,
|
| 28 |
-
"learning_rate": 3e-05,
|
| 29 |
-
"loss": 1.5953,
|
| 30 |
-
"step": 3
|
| 31 |
-
},
|
| 32 |
-
{
|
| 33 |
-
"epoch": 0.1111111111111111,
|
| 34 |
-
"grad_norm": 15.9970121383667,
|
| 35 |
-
"learning_rate": 4e-05,
|
| 36 |
-
"loss": 1.5355,
|
| 37 |
-
"step": 4
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"epoch": 0.1388888888888889,
|
| 41 |
-
"grad_norm": 18.634761810302734,
|
| 42 |
-
"learning_rate": 5e-05,
|
| 43 |
-
"loss": 1.552,
|
| 44 |
-
"step": 5
|
| 45 |
-
},
|
| 46 |
-
{
|
| 47 |
-
"epoch": 0.16666666666666666,
|
| 48 |
-
"grad_norm": 10.22042179107666,
|
| 49 |
-
"learning_rate": 6e-05,
|
| 50 |
-
"loss": 1.4926,
|
| 51 |
-
"step": 6
|
| 52 |
-
},
|
| 53 |
-
{
|
| 54 |
-
"epoch": 0.19444444444444445,
|
| 55 |
-
"grad_norm": 14.976595878601074,
|
| 56 |
-
"learning_rate": 7e-05,
|
| 57 |
-
"loss": 1.1827,
|
| 58 |
-
"step": 7
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"epoch": 0.2222222222222222,
|
| 62 |
-
"grad_norm": 34.35334396362305,
|
| 63 |
-
"learning_rate": 8e-05,
|
| 64 |
-
"loss": 1.173,
|
| 65 |
-
"step": 8
|
| 66 |
-
},
|
| 67 |
-
{
|
| 68 |
-
"epoch": 0.25,
|
| 69 |
-
"grad_norm": 7.392702579498291,
|
| 70 |
-
"learning_rate": 9e-05,
|
| 71 |
-
"loss": 0.7987,
|
| 72 |
-
"step": 9
|
| 73 |
-
},
|
| 74 |
-
{
|
| 75 |
-
"epoch": 0.2777777777777778,
|
| 76 |
-
"grad_norm": 8.6481294631958,
|
| 77 |
-
"learning_rate": 0.0001,
|
| 78 |
-
"loss": 0.7641,
|
| 79 |
-
"step": 10
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.2777777777777778,
|
| 83 |
-
"eval_loss": 0.719104528427124,
|
| 84 |
-
"eval_runtime": 0.1041,
|
| 85 |
-
"eval_samples_per_second": 441.967,
|
| 86 |
-
"eval_steps_per_second": 9.608,
|
| 87 |
-
"step": 10
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.3055555555555556,
|
| 91 |
-
"grad_norm": 11.896777153015137,
|
| 92 |
-
"learning_rate": 9.971428571428571e-05,
|
| 93 |
-
"loss": 0.6656,
|
| 94 |
-
"step": 11
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.3333333333333333,
|
| 98 |
-
"grad_norm": 6.002552032470703,
|
| 99 |
-
"learning_rate": 9.942857142857144e-05,
|
| 100 |
-
"loss": 0.5743,
|
| 101 |
-
"step": 12
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.3611111111111111,
|
| 105 |
-
"grad_norm": 4.569210529327393,
|
| 106 |
-
"learning_rate": 9.914285714285715e-05,
|
| 107 |
-
"loss": 0.5204,
|
| 108 |
-
"step": 13
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.3888888888888889,
|
| 112 |
-
"grad_norm": 3.2792978286743164,
|
| 113 |
-
"learning_rate": 9.885714285714286e-05,
|
| 114 |
-
"loss": 0.5217,
|
| 115 |
-
"step": 14
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.4166666666666667,
|
| 119 |
-
"grad_norm": 3.6701369285583496,
|
| 120 |
-
"learning_rate": 9.857142857142858e-05,
|
| 121 |
-
"loss": 0.5046,
|
| 122 |
-
"step": 15
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.4444444444444444,
|
| 126 |
-
"grad_norm": 4.064358711242676,
|
| 127 |
-
"learning_rate": 9.828571428571429e-05,
|
| 128 |
-
"loss": 0.4803,
|
| 129 |
-
"step": 16
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.4722222222222222,
|
| 133 |
-
"grad_norm": 2.9059529304504395,
|
| 134 |
-
"learning_rate": 9.8e-05,
|
| 135 |
-
"loss": 0.4466,
|
| 136 |
-
"step": 17
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.5,
|
| 140 |
-
"grad_norm": 2.499434471130371,
|
| 141 |
-
"learning_rate": 9.771428571428572e-05,
|
| 142 |
-
"loss": 0.4453,
|
| 143 |
-
"step": 18
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.5277777777777778,
|
| 147 |
-
"grad_norm": 2.084019899368286,
|
| 148 |
-
"learning_rate": 9.742857142857143e-05,
|
| 149 |
-
"loss": 0.4382,
|
| 150 |
-
"step": 19
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 0.5555555555555556,
|
| 154 |
-
"grad_norm": 1.2787052392959595,
|
| 155 |
-
"learning_rate": 9.714285714285715e-05,
|
| 156 |
-
"loss": 0.4208,
|
| 157 |
-
"step": 20
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 0.5555555555555556,
|
| 161 |
-
"eval_loss": 0.4429771900177002,
|
| 162 |
-
"eval_runtime": 0.1036,
|
| 163 |
-
"eval_samples_per_second": 443.983,
|
| 164 |
-
"eval_steps_per_second": 9.652,
|
| 165 |
-
"step": 20
|
| 166 |
-
},
|
| 167 |
-
{
|
| 168 |
-
"epoch": 0.5833333333333334,
|
| 169 |
-
"grad_norm": 1.8704657554626465,
|
| 170 |
-
"learning_rate": 9.685714285714286e-05,
|
| 171 |
-
"loss": 0.431,
|
| 172 |
-
"step": 21
|
| 173 |
-
},
|
| 174 |
-
{
|
| 175 |
-
"epoch": 0.6111111111111112,
|
| 176 |
-
"grad_norm": 1.7761015892028809,
|
| 177 |
-
"learning_rate": 9.657142857142858e-05,
|
| 178 |
-
"loss": 0.4317,
|
| 179 |
-
"step": 22
|
| 180 |
-
},
|
| 181 |
-
{
|
| 182 |
-
"epoch": 0.6388888888888888,
|
| 183 |
-
"grad_norm": 1.7499357461929321,
|
| 184 |
-
"learning_rate": 9.628571428571429e-05,
|
| 185 |
-
"loss": 0.4229,
|
| 186 |
-
"step": 23
|
| 187 |
-
},
|
| 188 |
-
{
|
| 189 |
-
"epoch": 0.6666666666666666,
|
| 190 |
-
"grad_norm": 1.2509069442749023,
|
| 191 |
-
"learning_rate": 9.6e-05,
|
| 192 |
-
"loss": 0.4126,
|
| 193 |
-
"step": 24
|
| 194 |
-
},
|
| 195 |
-
{
|
| 196 |
-
"epoch": 0.6944444444444444,
|
| 197 |
-
"grad_norm": 1.3415058851242065,
|
| 198 |
-
"learning_rate": 9.571428571428573e-05,
|
| 199 |
-
"loss": 0.3888,
|
| 200 |
-
"step": 25
|
| 201 |
-
},
|
| 202 |
-
{
|
| 203 |
-
"epoch": 0.7222222222222222,
|
| 204 |
-
"grad_norm": 1.513482689857483,
|
| 205 |
-
"learning_rate": 9.542857142857143e-05,
|
| 206 |
-
"loss": 0.4111,
|
| 207 |
-
"step": 26
|
| 208 |
-
},
|
| 209 |
-
{
|
| 210 |
-
"epoch": 0.75,
|
| 211 |
-
"grad_norm": 1.0207685232162476,
|
| 212 |
-
"learning_rate": 9.514285714285714e-05,
|
| 213 |
-
"loss": 0.3904,
|
| 214 |
-
"step": 27
|
| 215 |
-
},
|
| 216 |
-
{
|
| 217 |
-
"epoch": 0.7777777777777778,
|
| 218 |
-
"grad_norm": 1.0765091180801392,
|
| 219 |
-
"learning_rate": 9.485714285714287e-05,
|
| 220 |
-
"loss": 0.3911,
|
| 221 |
-
"step": 28
|
| 222 |
-
},
|
| 223 |
-
{
|
| 224 |
-
"epoch": 0.8055555555555556,
|
| 225 |
-
"grad_norm": 1.2146029472351074,
|
| 226 |
-
"learning_rate": 9.457142857142858e-05,
|
| 227 |
-
"loss": 0.3893,
|
| 228 |
-
"step": 29
|
| 229 |
-
},
|
| 230 |
-
{
|
| 231 |
-
"epoch": 0.8333333333333334,
|
| 232 |
-
"grad_norm": 1.302972435951233,
|
| 233 |
-
"learning_rate": 9.428571428571429e-05,
|
| 234 |
-
"loss": 0.3915,
|
| 235 |
-
"step": 30
|
| 236 |
-
},
|
| 237 |
-
{
|
| 238 |
-
"epoch": 0.8333333333333334,
|
| 239 |
-
"eval_loss": 0.39956018328666687,
|
| 240 |
-
"eval_runtime": 0.1036,
|
| 241 |
-
"eval_samples_per_second": 444.089,
|
| 242 |
-
"eval_steps_per_second": 9.654,
|
| 243 |
-
"step": 30
|
| 244 |
-
},
|
| 245 |
-
{
|
| 246 |
-
"epoch": 0.8611111111111112,
|
| 247 |
-
"grad_norm": 1.2195433378219604,
|
| 248 |
-
"learning_rate": 9.4e-05,
|
| 249 |
-
"loss": 0.377,
|
| 250 |
-
"step": 31
|
| 251 |
-
},
|
| 252 |
-
{
|
| 253 |
-
"epoch": 0.8888888888888888,
|
| 254 |
-
"grad_norm": 1.2320094108581543,
|
| 255 |
-
"learning_rate": 9.371428571428572e-05,
|
| 256 |
-
"loss": 0.3837,
|
| 257 |
-
"step": 32
|
| 258 |
-
},
|
| 259 |
-
{
|
| 260 |
-
"epoch": 0.9166666666666666,
|
| 261 |
-
"grad_norm": 1.0609043836593628,
|
| 262 |
-
"learning_rate": 9.342857142857143e-05,
|
| 263 |
-
"loss": 0.3776,
|
| 264 |
-
"step": 33
|
| 265 |
-
},
|
| 266 |
-
{
|
| 267 |
-
"epoch": 0.9444444444444444,
|
| 268 |
-
"grad_norm": 0.9609966278076172,
|
| 269 |
-
"learning_rate": 9.314285714285715e-05,
|
| 270 |
-
"loss": 0.3887,
|
| 271 |
-
"step": 34
|
| 272 |
-
},
|
| 273 |
-
{
|
| 274 |
-
"epoch": 0.9722222222222222,
|
| 275 |
-
"grad_norm": 1.0595581531524658,
|
| 276 |
-
"learning_rate": 9.285714285714286e-05,
|
| 277 |
-
"loss": 0.3761,
|
| 278 |
-
"step": 35
|
| 279 |
-
},
|
| 280 |
-
{
|
| 281 |
-
"epoch": 1.0,
|
| 282 |
-
"grad_norm": 0.990327775478363,
|
| 283 |
-
"learning_rate": 9.257142857142858e-05,
|
| 284 |
-
"loss": 0.3744,
|
| 285 |
-
"step": 36
|
| 286 |
-
},
|
| 287 |
-
{
|
| 288 |
-
"epoch": 1.0277777777777777,
|
| 289 |
-
"grad_norm": 1.272873044013977,
|
| 290 |
-
"learning_rate": 9.228571428571429e-05,
|
| 291 |
-
"loss": 0.3625,
|
| 292 |
-
"step": 37
|
| 293 |
-
},
|
| 294 |
-
{
|
| 295 |
-
"epoch": 1.0555555555555556,
|
| 296 |
-
"grad_norm": 1.9024567604064941,
|
| 297 |
-
"learning_rate": 9.200000000000001e-05,
|
| 298 |
-
"loss": 0.3869,
|
| 299 |
-
"step": 38
|
| 300 |
-
},
|
| 301 |
-
{
|
| 302 |
-
"epoch": 1.0833333333333333,
|
| 303 |
-
"grad_norm": 1.3398654460906982,
|
| 304 |
-
"learning_rate": 9.171428571428572e-05,
|
| 305 |
-
"loss": 0.3751,
|
| 306 |
-
"step": 39
|
| 307 |
-
},
|
| 308 |
-
{
|
| 309 |
-
"epoch": 1.1111111111111112,
|
| 310 |
-
"grad_norm": 1.9176064729690552,
|
| 311 |
-
"learning_rate": 9.142857142857143e-05,
|
| 312 |
-
"loss": 0.3662,
|
| 313 |
-
"step": 40
|
| 314 |
-
},
|
| 315 |
-
{
|
| 316 |
-
"epoch": 1.1111111111111112,
|
| 317 |
-
"eval_loss": 0.37765416502952576,
|
| 318 |
-
"eval_runtime": 0.1036,
|
| 319 |
-
"eval_samples_per_second": 444.001,
|
| 320 |
-
"eval_steps_per_second": 9.652,
|
| 321 |
-
"step": 40
|
| 322 |
-
},
|
| 323 |
-
{
|
| 324 |
-
"epoch": 1.1388888888888888,
|
| 325 |
-
"grad_norm": 1.1852660179138184,
|
| 326 |
-
"learning_rate": 9.114285714285716e-05,
|
| 327 |
-
"loss": 0.3665,
|
| 328 |
-
"step": 41
|
| 329 |
-
},
|
| 330 |
-
{
|
| 331 |
-
"epoch": 1.1666666666666667,
|
| 332 |
-
"grad_norm": 1.831186056137085,
|
| 333 |
-
"learning_rate": 9.085714285714286e-05,
|
| 334 |
-
"loss": 0.3705,
|
| 335 |
-
"step": 42
|
| 336 |
-
},
|
| 337 |
-
{
|
| 338 |
-
"epoch": 1.1944444444444444,
|
| 339 |
-
"grad_norm": 1.1574777364730835,
|
| 340 |
-
"learning_rate": 9.057142857142857e-05,
|
| 341 |
-
"loss": 0.3582,
|
| 342 |
-
"step": 43
|
| 343 |
-
},
|
| 344 |
-
{
|
| 345 |
-
"epoch": 1.2222222222222223,
|
| 346 |
-
"grad_norm": 1.3485198020935059,
|
| 347 |
-
"learning_rate": 9.028571428571428e-05,
|
| 348 |
-
"loss": 0.3724,
|
| 349 |
-
"step": 44
|
| 350 |
-
},
|
| 351 |
-
{
|
| 352 |
-
"epoch": 1.25,
|
| 353 |
-
"grad_norm": 1.0934721231460571,
|
| 354 |
-
"learning_rate": 9e-05,
|
| 355 |
-
"loss": 0.3549,
|
| 356 |
-
"step": 45
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 1.2777777777777777,
|
| 360 |
-
"grad_norm": 1.2588518857955933,
|
| 361 |
-
"learning_rate": 8.971428571428571e-05,
|
| 362 |
-
"loss": 0.3607,
|
| 363 |
-
"step": 46
|
| 364 |
-
},
|
| 365 |
-
{
|
| 366 |
-
"epoch": 1.3055555555555556,
|
| 367 |
-
"grad_norm": 0.9038533568382263,
|
| 368 |
-
"learning_rate": 8.942857142857142e-05,
|
| 369 |
-
"loss": 0.3492,
|
| 370 |
-
"step": 47
|
| 371 |
-
},
|
| 372 |
-
{
|
| 373 |
-
"epoch": 1.3333333333333333,
|
| 374 |
-
"grad_norm": 1.083348274230957,
|
| 375 |
-
"learning_rate": 8.914285714285715e-05,
|
| 376 |
-
"loss": 0.361,
|
| 377 |
-
"step": 48
|
| 378 |
-
},
|
| 379 |
-
{
|
| 380 |
-
"epoch": 1.3611111111111112,
|
| 381 |
-
"grad_norm": 0.8287424445152283,
|
| 382 |
-
"learning_rate": 8.885714285714286e-05,
|
| 383 |
-
"loss": 0.3475,
|
| 384 |
-
"step": 49
|
| 385 |
-
},
|
| 386 |
-
{
|
| 387 |
-
"epoch": 1.3888888888888888,
|
| 388 |
-
"grad_norm": 1.3475714921951294,
|
| 389 |
-
"learning_rate": 8.857142857142857e-05,
|
| 390 |
-
"loss": 0.363,
|
| 391 |
-
"step": 50
|
| 392 |
-
},
|
| 393 |
-
{
|
| 394 |
-
"epoch": 1.3888888888888888,
|
| 395 |
-
"eval_loss": 0.3627206087112427,
|
| 396 |
-
"eval_runtime": 0.1039,
|
| 397 |
-
"eval_samples_per_second": 442.561,
|
| 398 |
-
"eval_steps_per_second": 9.621,
|
| 399 |
-
"step": 50
|
| 400 |
-
},
|
| 401 |
-
{
|
| 402 |
-
"epoch": 1.4166666666666667,
|
| 403 |
-
"grad_norm": 1.1012552976608276,
|
| 404 |
-
"learning_rate": 8.828571428571429e-05,
|
| 405 |
-
"loss": 0.3527,
|
| 406 |
-
"step": 51
|
| 407 |
-
},
|
| 408 |
-
{
|
| 409 |
-
"epoch": 1.4444444444444444,
|
| 410 |
-
"grad_norm": 0.6421935558319092,
|
| 411 |
-
"learning_rate": 8.800000000000001e-05,
|
| 412 |
-
"loss": 0.3418,
|
| 413 |
-
"step": 52
|
| 414 |
-
},
|
| 415 |
-
{
|
| 416 |
-
"epoch": 1.4722222222222223,
|
| 417 |
-
"grad_norm": 1.1574995517730713,
|
| 418 |
-
"learning_rate": 8.771428571428572e-05,
|
| 419 |
-
"loss": 0.3513,
|
| 420 |
-
"step": 53
|
| 421 |
-
},
|
| 422 |
-
{
|
| 423 |
-
"epoch": 1.5,
|
| 424 |
-
"grad_norm": 1.0251258611679077,
|
| 425 |
-
"learning_rate": 8.742857142857144e-05,
|
| 426 |
-
"loss": 0.3515,
|
| 427 |
-
"step": 54
|
| 428 |
-
},
|
| 429 |
-
{
|
| 430 |
-
"epoch": 1.5277777777777777,
|
| 431 |
-
"grad_norm": 0.9864039421081543,
|
| 432 |
-
"learning_rate": 8.714285714285715e-05,
|
| 433 |
-
"loss": 0.3609,
|
| 434 |
-
"step": 55
|
| 435 |
-
},
|
| 436 |
-
{
|
| 437 |
-
"epoch": 1.5555555555555556,
|
| 438 |
-
"grad_norm": 0.757999062538147,
|
| 439 |
-
"learning_rate": 8.685714285714286e-05,
|
| 440 |
-
"loss": 0.3454,
|
| 441 |
-
"step": 56
|
| 442 |
-
},
|
| 443 |
-
{
|
| 444 |
-
"epoch": 1.5833333333333335,
|
| 445 |
-
"grad_norm": 1.0983614921569824,
|
| 446 |
-
"learning_rate": 8.657142857142858e-05,
|
| 447 |
-
"loss": 0.3488,
|
| 448 |
-
"step": 57
|
| 449 |
-
},
|
| 450 |
-
{
|
| 451 |
-
"epoch": 1.6111111111111112,
|
| 452 |
-
"grad_norm": 1.4811136722564697,
|
| 453 |
-
"learning_rate": 8.62857142857143e-05,
|
| 454 |
-
"loss": 0.3562,
|
| 455 |
-
"step": 58
|
| 456 |
-
},
|
| 457 |
-
{
|
| 458 |
-
"epoch": 1.6388888888888888,
|
| 459 |
-
"grad_norm": 0.9457672834396362,
|
| 460 |
-
"learning_rate": 8.6e-05,
|
| 461 |
-
"loss": 0.349,
|
| 462 |
-
"step": 59
|
| 463 |
-
},
|
| 464 |
-
{
|
| 465 |
-
"epoch": 1.6666666666666665,
|
| 466 |
-
"grad_norm": 1.4347460269927979,
|
| 467 |
-
"learning_rate": 8.571428571428571e-05,
|
| 468 |
-
"loss": 0.3551,
|
| 469 |
-
"step": 60
|
| 470 |
-
},
|
| 471 |
-
{
|
| 472 |
-
"epoch": 1.6666666666666665,
|
| 473 |
-
"eval_loss": 0.35965070128440857,
|
| 474 |
-
"eval_runtime": 0.1045,
|
| 475 |
-
"eval_samples_per_second": 440.164,
|
| 476 |
-
"eval_steps_per_second": 9.569,
|
| 477 |
-
"step": 60
|
| 478 |
-
},
|
| 479 |
-
{
|
| 480 |
-
"epoch": 1.6944444444444444,
|
| 481 |
-
"grad_norm": 1.0592706203460693,
|
| 482 |
-
"learning_rate": 8.542857142857144e-05,
|
| 483 |
-
"loss": 0.3485,
|
| 484 |
-
"step": 61
|
| 485 |
-
},
|
| 486 |
-
{
|
| 487 |
-
"epoch": 1.7222222222222223,
|
| 488 |
-
"grad_norm": 1.3444126844406128,
|
| 489 |
-
"learning_rate": 8.514285714285714e-05,
|
| 490 |
-
"loss": 0.3512,
|
| 491 |
-
"step": 62
|
| 492 |
-
},
|
| 493 |
-
{
|
| 494 |
-
"epoch": 1.75,
|
| 495 |
-
"grad_norm": 0.9045667052268982,
|
| 496 |
-
"learning_rate": 8.485714285714285e-05,
|
| 497 |
-
"loss": 0.3525,
|
| 498 |
-
"step": 63
|
| 499 |
-
},
|
| 500 |
-
{
|
| 501 |
-
"epoch": 1.7777777777777777,
|
| 502 |
-
"grad_norm": 1.135429859161377,
|
| 503 |
-
"learning_rate": 8.457142857142858e-05,
|
| 504 |
-
"loss": 0.3483,
|
| 505 |
-
"step": 64
|
| 506 |
-
},
|
| 507 |
-
{
|
| 508 |
-
"epoch": 1.8055555555555556,
|
| 509 |
-
"grad_norm": 0.7742411494255066,
|
| 510 |
-
"learning_rate": 8.428571428571429e-05,
|
| 511 |
-
"loss": 0.3445,
|
| 512 |
-
"step": 65
|
| 513 |
-
},
|
| 514 |
-
{
|
| 515 |
-
"epoch": 1.8333333333333335,
|
| 516 |
-
"grad_norm": 1.2747840881347656,
|
| 517 |
-
"learning_rate": 8.4e-05,
|
| 518 |
-
"loss": 0.3425,
|
| 519 |
-
"step": 66
|
| 520 |
-
},
|
| 521 |
-
{
|
| 522 |
-
"epoch": 1.8611111111111112,
|
| 523 |
-
"grad_norm": 1.1280975341796875,
|
| 524 |
-
"learning_rate": 8.371428571428572e-05,
|
| 525 |
-
"loss": 0.3506,
|
| 526 |
-
"step": 67
|
| 527 |
-
},
|
| 528 |
-
{
|
| 529 |
-
"epoch": 1.8888888888888888,
|
| 530 |
-
"grad_norm": 1.3229925632476807,
|
| 531 |
-
"learning_rate": 8.342857142857143e-05,
|
| 532 |
-
"loss": 0.3458,
|
| 533 |
-
"step": 68
|
| 534 |
-
},
|
| 535 |
-
{
|
| 536 |
-
"epoch": 1.9166666666666665,
|
| 537 |
-
"grad_norm": 1.0970568656921387,
|
| 538 |
-
"learning_rate": 8.314285714285715e-05,
|
| 539 |
-
"loss": 0.3443,
|
| 540 |
-
"step": 69
|
| 541 |
-
},
|
| 542 |
-
{
|
| 543 |
-
"epoch": 1.9444444444444444,
|
| 544 |
-
"grad_norm": 1.7599389553070068,
|
| 545 |
-
"learning_rate": 8.285714285714287e-05,
|
| 546 |
-
"loss": 0.3612,
|
| 547 |
-
"step": 70
|
| 548 |
-
},
|
| 549 |
-
{
|
| 550 |
-
"epoch": 1.9444444444444444,
|
| 551 |
-
"eval_loss": 0.35299769043922424,
|
| 552 |
-
"eval_runtime": 0.1042,
|
| 553 |
-
"eval_samples_per_second": 441.433,
|
| 554 |
-
"eval_steps_per_second": 9.596,
|
| 555 |
-
"step": 70
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 1.9722222222222223,
|
| 559 |
-
"grad_norm": 0.8275991678237915,
|
| 560 |
-
"learning_rate": 8.257142857142858e-05,
|
| 561 |
-
"loss": 0.3373,
|
| 562 |
-
"step": 71
|
| 563 |
-
},
|
| 564 |
-
{
|
| 565 |
-
"epoch": 2.0,
|
| 566 |
-
"grad_norm": 1.5045437812805176,
|
| 567 |
-
"learning_rate": 8.228571428571429e-05,
|
| 568 |
-
"loss": 0.3624,
|
| 569 |
-
"step": 72
|
| 570 |
-
},
|
| 571 |
-
{
|
| 572 |
-
"epoch": 2.0277777777777777,
|
| 573 |
-
"grad_norm": 0.9771829843521118,
|
| 574 |
-
"learning_rate": 8.2e-05,
|
| 575 |
-
"loss": 0.3434,
|
| 576 |
-
"step": 73
|
| 577 |
-
},
|
| 578 |
-
{
|
| 579 |
-
"epoch": 2.0555555555555554,
|
| 580 |
-
"grad_norm": 0.8552800416946411,
|
| 581 |
-
"learning_rate": 8.171428571428572e-05,
|
| 582 |
-
"loss": 0.3347,
|
| 583 |
-
"step": 74
|
| 584 |
-
},
|
| 585 |
-
{
|
| 586 |
-
"epoch": 2.0833333333333335,
|
| 587 |
-
"grad_norm": 0.8917291164398193,
|
| 588 |
-
"learning_rate": 8.142857142857143e-05,
|
| 589 |
-
"loss": 0.3448,
|
| 590 |
-
"step": 75
|
| 591 |
-
},
|
| 592 |
-
{
|
| 593 |
-
"epoch": 2.111111111111111,
|
| 594 |
-
"grad_norm": 0.9143850207328796,
|
| 595 |
-
"learning_rate": 8.114285714285714e-05,
|
| 596 |
-
"loss": 0.3309,
|
| 597 |
-
"step": 76
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 2.138888888888889,
|
| 601 |
-
"grad_norm": 1.359926700592041,
|
| 602 |
-
"learning_rate": 8.085714285714287e-05,
|
| 603 |
-
"loss": 0.3471,
|
| 604 |
-
"step": 77
|
| 605 |
-
},
|
| 606 |
-
{
|
| 607 |
-
"epoch": 2.1666666666666665,
|
| 608 |
-
"grad_norm": 0.84107506275177,
|
| 609 |
-
"learning_rate": 8.057142857142857e-05,
|
| 610 |
-
"loss": 0.3433,
|
| 611 |
-
"step": 78
|
| 612 |
-
},
|
| 613 |
-
{
|
| 614 |
-
"epoch": 2.1944444444444446,
|
| 615 |
-
"grad_norm": 1.2953639030456543,
|
| 616 |
-
"learning_rate": 8.028571428571428e-05,
|
| 617 |
-
"loss": 0.3495,
|
| 618 |
-
"step": 79
|
| 619 |
-
},
|
| 620 |
-
{
|
| 621 |
-
"epoch": 2.2222222222222223,
|
| 622 |
-
"grad_norm": 0.9937311410903931,
|
| 623 |
-
"learning_rate": 8e-05,
|
| 624 |
-
"loss": 0.3388,
|
| 625 |
-
"step": 80
|
| 626 |
-
},
|
| 627 |
-
{
|
| 628 |
-
"epoch": 2.2222222222222223,
|
| 629 |
-
"eval_loss": 0.3491460978984833,
|
| 630 |
-
"eval_runtime": 0.1044,
|
| 631 |
-
"eval_samples_per_second": 440.593,
|
| 632 |
-
"eval_steps_per_second": 9.578,
|
| 633 |
-
"step": 80
|
| 634 |
-
},
|
| 635 |
-
{
|
| 636 |
-
"epoch": 2.25,
|
| 637 |
-
"grad_norm": 1.0681490898132324,
|
| 638 |
-
"learning_rate": 7.971428571428572e-05,
|
| 639 |
-
"loss": 0.3435,
|
| 640 |
-
"step": 81
|
| 641 |
-
},
|
| 642 |
-
{
|
| 643 |
-
"epoch": 2.2777777777777777,
|
| 644 |
-
"grad_norm": 0.8466928601264954,
|
| 645 |
-
"learning_rate": 7.942857142857143e-05,
|
| 646 |
-
"loss": 0.3333,
|
| 647 |
-
"step": 82
|
| 648 |
-
},
|
| 649 |
-
{
|
| 650 |
-
"epoch": 2.3055555555555554,
|
| 651 |
-
"grad_norm": 0.8183342814445496,
|
| 652 |
-
"learning_rate": 7.914285714285715e-05,
|
| 653 |
-
"loss": 0.3305,
|
| 654 |
-
"step": 83
|
| 655 |
-
},
|
| 656 |
-
{
|
| 657 |
-
"epoch": 2.3333333333333335,
|
| 658 |
-
"grad_norm": 0.833314061164856,
|
| 659 |
-
"learning_rate": 7.885714285714286e-05,
|
| 660 |
-
"loss": 0.3289,
|
| 661 |
-
"step": 84
|
| 662 |
-
},
|
| 663 |
-
{
|
| 664 |
-
"epoch": 2.361111111111111,
|
| 665 |
-
"grad_norm": 0.8347731828689575,
|
| 666 |
-
"learning_rate": 7.857142857142858e-05,
|
| 667 |
-
"loss": 0.3331,
|
| 668 |
-
"step": 85
|
| 669 |
-
},
|
| 670 |
-
{
|
| 671 |
-
"epoch": 2.388888888888889,
|
| 672 |
-
"grad_norm": 1.0877679586410522,
|
| 673 |
-
"learning_rate": 7.828571428571429e-05,
|
| 674 |
-
"loss": 0.3437,
|
| 675 |
-
"step": 86
|
| 676 |
-
},
|
| 677 |
-
{
|
| 678 |
-
"epoch": 2.4166666666666665,
|
| 679 |
-
"grad_norm": 0.9570125937461853,
|
| 680 |
-
"learning_rate": 7.800000000000001e-05,
|
| 681 |
-
"loss": 0.3331,
|
| 682 |
-
"step": 87
|
| 683 |
-
},
|
| 684 |
-
{
|
| 685 |
-
"epoch": 2.4444444444444446,
|
| 686 |
-
"grad_norm": 0.7662280797958374,
|
| 687 |
-
"learning_rate": 7.771428571428572e-05,
|
| 688 |
-
"loss": 0.3363,
|
| 689 |
-
"step": 88
|
| 690 |
-
},
|
| 691 |
-
{
|
| 692 |
-
"epoch": 2.4722222222222223,
|
| 693 |
-
"grad_norm": 0.9321999549865723,
|
| 694 |
-
"learning_rate": 7.742857142857143e-05,
|
| 695 |
-
"loss": 0.3305,
|
| 696 |
-
"step": 89
|
| 697 |
-
},
|
| 698 |
-
{
|
| 699 |
-
"epoch": 2.5,
|
| 700 |
-
"grad_norm": 0.8284544348716736,
|
| 701 |
-
"learning_rate": 7.714285714285715e-05,
|
| 702 |
-
"loss": 0.3332,
|
| 703 |
-
"step": 90
|
| 704 |
-
},
|
| 705 |
-
{
|
| 706 |
-
"epoch": 2.5,
|
| 707 |
-
"eval_loss": 0.34498000144958496,
|
| 708 |
-
"eval_runtime": 0.1046,
|
| 709 |
-
"eval_samples_per_second": 439.848,
|
| 710 |
-
"eval_steps_per_second": 9.562,
|
| 711 |
-
"step": 90
|
| 712 |
-
},
|
| 713 |
-
{
|
| 714 |
-
"epoch": 2.5277777777777777,
|
| 715 |
-
"grad_norm": 1.0568827390670776,
|
| 716 |
-
"learning_rate": 7.685714285714286e-05,
|
| 717 |
-
"loss": 0.3448,
|
| 718 |
-
"step": 91
|
| 719 |
-
},
|
| 720 |
-
{
|
| 721 |
-
"epoch": 2.5555555555555554,
|
| 722 |
-
"grad_norm": 0.9136806130409241,
|
| 723 |
-
"learning_rate": 7.657142857142857e-05,
|
| 724 |
-
"loss": 0.334,
|
| 725 |
-
"step": 92
|
| 726 |
-
},
|
| 727 |
-
{
|
| 728 |
-
"epoch": 2.5833333333333335,
|
| 729 |
-
"grad_norm": 1.2551990747451782,
|
| 730 |
-
"learning_rate": 7.62857142857143e-05,
|
| 731 |
-
"loss": 0.3425,
|
| 732 |
-
"step": 93
|
| 733 |
-
},
|
| 734 |
-
{
|
| 735 |
-
"epoch": 2.611111111111111,
|
| 736 |
-
"grad_norm": 0.8284862637519836,
|
| 737 |
-
"learning_rate": 7.6e-05,
|
| 738 |
-
"loss": 0.3265,
|
| 739 |
-
"step": 94
|
| 740 |
-
},
|
| 741 |
-
{
|
| 742 |
-
"epoch": 2.638888888888889,
|
| 743 |
-
"grad_norm": 0.7161554098129272,
|
| 744 |
-
"learning_rate": 7.571428571428571e-05,
|
| 745 |
-
"loss": 0.3267,
|
| 746 |
-
"step": 95
|
| 747 |
-
},
|
| 748 |
-
{
|
| 749 |
-
"epoch": 2.6666666666666665,
|
| 750 |
-
"grad_norm": 0.8050905466079712,
|
| 751 |
-
"learning_rate": 7.542857142857144e-05,
|
| 752 |
-
"loss": 0.3342,
|
| 753 |
-
"step": 96
|
| 754 |
-
},
|
| 755 |
-
{
|
| 756 |
-
"epoch": 2.6944444444444446,
|
| 757 |
-
"grad_norm": 0.7441209554672241,
|
| 758 |
-
"learning_rate": 7.514285714285715e-05,
|
| 759 |
-
"loss": 0.3325,
|
| 760 |
-
"step": 97
|
| 761 |
-
},
|
| 762 |
-
{
|
| 763 |
-
"epoch": 2.7222222222222223,
|
| 764 |
-
"grad_norm": 0.591927707195282,
|
| 765 |
-
"learning_rate": 7.485714285714285e-05,
|
| 766 |
-
"loss": 0.334,
|
| 767 |
-
"step": 98
|
| 768 |
-
},
|
| 769 |
-
{
|
| 770 |
-
"epoch": 2.75,
|
| 771 |
-
"grad_norm": 0.8902866244316101,
|
| 772 |
-
"learning_rate": 7.457142857142856e-05,
|
| 773 |
-
"loss": 0.3473,
|
| 774 |
-
"step": 99
|
| 775 |
-
},
|
| 776 |
-
{
|
| 777 |
-
"epoch": 2.7777777777777777,
|
| 778 |
-
"grad_norm": 0.6760069131851196,
|
| 779 |
-
"learning_rate": 7.428571428571429e-05,
|
| 780 |
-
"loss": 0.326,
|
| 781 |
-
"step": 100
|
| 782 |
-
},
|
| 783 |
-
{
|
| 784 |
-
"epoch": 2.7777777777777777,
|
| 785 |
-
"eval_loss": 0.3410107493400574,
|
| 786 |
-
"eval_runtime": 0.1043,
|
| 787 |
-
"eval_samples_per_second": 441.026,
|
| 788 |
-
"eval_steps_per_second": 9.588,
|
| 789 |
-
"step": 100
|
| 790 |
-
},
|
| 791 |
-
{
|
| 792 |
-
"epoch": 2.8055555555555554,
|
| 793 |
-
"grad_norm": 0.6579345464706421,
|
| 794 |
-
"learning_rate": 7.4e-05,
|
| 795 |
-
"loss": 0.3348,
|
| 796 |
-
"step": 101
|
| 797 |
-
},
|
| 798 |
-
{
|
| 799 |
-
"epoch": 2.8333333333333335,
|
| 800 |
-
"grad_norm": 1.0648415088653564,
|
| 801 |
-
"learning_rate": 7.371428571428572e-05,
|
| 802 |
-
"loss": 0.3253,
|
| 803 |
-
"step": 102
|
| 804 |
-
},
|
| 805 |
-
{
|
| 806 |
-
"epoch": 2.861111111111111,
|
| 807 |
-
"grad_norm": 0.6868814826011658,
|
| 808 |
-
"learning_rate": 7.342857142857144e-05,
|
| 809 |
-
"loss": 0.3297,
|
| 810 |
-
"step": 103
|
| 811 |
-
},
|
| 812 |
-
{
|
| 813 |
-
"epoch": 2.888888888888889,
|
| 814 |
-
"grad_norm": 1.1149464845657349,
|
| 815 |
-
"learning_rate": 7.314285714285715e-05,
|
| 816 |
-
"loss": 0.3401,
|
| 817 |
-
"step": 104
|
| 818 |
-
},
|
| 819 |
-
{
|
| 820 |
-
"epoch": 2.9166666666666665,
|
| 821 |
-
"grad_norm": 0.8934164047241211,
|
| 822 |
-
"learning_rate": 7.285714285714286e-05,
|
| 823 |
-
"loss": 0.3348,
|
| 824 |
-
"step": 105
|
| 825 |
-
},
|
| 826 |
-
{
|
| 827 |
-
"epoch": 2.9444444444444446,
|
| 828 |
-
"grad_norm": 1.1119507551193237,
|
| 829 |
-
"learning_rate": 7.257142857142858e-05,
|
| 830 |
-
"loss": 0.3427,
|
| 831 |
-
"step": 106
|
| 832 |
-
},
|
| 833 |
-
{
|
| 834 |
-
"epoch": 2.9722222222222223,
|
| 835 |
-
"grad_norm": 0.8103634715080261,
|
| 836 |
-
"learning_rate": 7.228571428571429e-05,
|
| 837 |
-
"loss": 0.3374,
|
| 838 |
-
"step": 107
|
| 839 |
-
},
|
| 840 |
-
{
|
| 841 |
-
"epoch": 3.0,
|
| 842 |
-
"grad_norm": 0.8421126008033752,
|
| 843 |
-
"learning_rate": 7.2e-05,
|
| 844 |
-
"loss": 0.3395,
|
| 845 |
-
"step": 108
|
| 846 |
-
},
|
| 847 |
-
{
|
| 848 |
-
"epoch": 3.0277777777777777,
|
| 849 |
-
"grad_norm": 0.8583278656005859,
|
| 850 |
-
"learning_rate": 7.171428571428572e-05,
|
| 851 |
-
"loss": 0.331,
|
| 852 |
-
"step": 109
|
| 853 |
-
},
|
| 854 |
-
{
|
| 855 |
-
"epoch": 3.0555555555555554,
|
| 856 |
-
"grad_norm": 1.2129111289978027,
|
| 857 |
-
"learning_rate": 7.142857142857143e-05,
|
| 858 |
-
"loss": 0.3355,
|
| 859 |
-
"step": 110
|
| 860 |
-
},
|
| 861 |
-
{
|
| 862 |
-
"epoch": 3.0555555555555554,
|
| 863 |
-
"eval_loss": 0.332057386636734,
|
| 864 |
-
"eval_runtime": 0.1051,
|
| 865 |
-
"eval_samples_per_second": 437.685,
|
| 866 |
-
"eval_steps_per_second": 9.515,
|
| 867 |
-
"step": 110
|
| 868 |
-
},
|
| 869 |
-
{
|
| 870 |
-
"epoch": 3.0833333333333335,
|
| 871 |
-
"grad_norm": 0.9463130235671997,
|
| 872 |
-
"learning_rate": 7.114285714285714e-05,
|
| 873 |
-
"loss": 0.3294,
|
| 874 |
-
"step": 111
|
| 875 |
-
},
|
| 876 |
-
{
|
| 877 |
-
"epoch": 3.111111111111111,
|
| 878 |
-
"grad_norm": 0.9692079424858093,
|
| 879 |
-
"learning_rate": 7.085714285714285e-05,
|
| 880 |
-
"loss": 0.3327,
|
| 881 |
-
"step": 112
|
| 882 |
-
},
|
| 883 |
-
{
|
| 884 |
-
"epoch": 3.138888888888889,
|
| 885 |
-
"grad_norm": 0.9853659868240356,
|
| 886 |
-
"learning_rate": 7.057142857142858e-05,
|
| 887 |
-
"loss": 0.3295,
|
| 888 |
-
"step": 113
|
| 889 |
-
},
|
| 890 |
-
{
|
| 891 |
-
"epoch": 3.1666666666666665,
|
| 892 |
-
"grad_norm": 0.7222715616226196,
|
| 893 |
-
"learning_rate": 7.028571428571428e-05,
|
| 894 |
-
"loss": 0.3358,
|
| 895 |
-
"step": 114
|
| 896 |
-
},
|
| 897 |
-
{
|
| 898 |
-
"epoch": 3.1944444444444446,
|
| 899 |
-
"grad_norm": 1.1528452634811401,
|
| 900 |
-
"learning_rate": 7e-05,
|
| 901 |
-
"loss": 0.3406,
|
| 902 |
-
"step": 115
|
| 903 |
-
},
|
| 904 |
-
{
|
| 905 |
-
"epoch": 3.2222222222222223,
|
| 906 |
-
"grad_norm": 1.0079970359802246,
|
| 907 |
-
"learning_rate": 6.971428571428572e-05,
|
| 908 |
-
"loss": 0.329,
|
| 909 |
-
"step": 116
|
| 910 |
-
},
|
| 911 |
-
{
|
| 912 |
-
"epoch": 3.25,
|
| 913 |
-
"grad_norm": 0.7162885665893555,
|
| 914 |
-
"learning_rate": 6.942857142857143e-05,
|
| 915 |
-
"loss": 0.327,
|
| 916 |
-
"step": 117
|
| 917 |
-
},
|
| 918 |
-
{
|
| 919 |
-
"epoch": 3.2777777777777777,
|
| 920 |
-
"grad_norm": 0.9302375912666321,
|
| 921 |
-
"learning_rate": 6.914285714285715e-05,
|
| 922 |
-
"loss": 0.336,
|
| 923 |
-
"step": 118
|
| 924 |
-
},
|
| 925 |
-
{
|
| 926 |
-
"epoch": 3.3055555555555554,
|
| 927 |
-
"grad_norm": 0.8540468215942383,
|
| 928 |
-
"learning_rate": 6.885714285714286e-05,
|
| 929 |
-
"loss": 0.3356,
|
| 930 |
-
"step": 119
|
| 931 |
-
},
|
| 932 |
-
{
|
| 933 |
-
"epoch": 3.3333333333333335,
|
| 934 |
-
"grad_norm": 0.598040759563446,
|
| 935 |
-
"learning_rate": 6.857142857142858e-05,
|
| 936 |
-
"loss": 0.3279,
|
| 937 |
-
"step": 120
|
| 938 |
-
},
|
| 939 |
-
{
|
| 940 |
-
"epoch": 3.3333333333333335,
|
| 941 |
-
"eval_loss": 0.3290035128593445,
|
| 942 |
-
"eval_runtime": 0.1049,
|
| 943 |
-
"eval_samples_per_second": 438.622,
|
| 944 |
-
"eval_steps_per_second": 9.535,
|
| 945 |
-
"step": 120
|
| 946 |
-
},
|
| 947 |
-
{
|
| 948 |
-
"epoch": 3.361111111111111,
|
| 949 |
-
"grad_norm": 0.6981043815612793,
|
| 950 |
-
"learning_rate": 6.828571428571429e-05,
|
| 951 |
-
"loss": 0.33,
|
| 952 |
-
"step": 121
|
| 953 |
-
},
|
| 954 |
-
{
|
| 955 |
-
"epoch": 3.388888888888889,
|
| 956 |
-
"grad_norm": 0.6710860133171082,
|
| 957 |
-
"learning_rate": 6.800000000000001e-05,
|
| 958 |
-
"loss": 0.3262,
|
| 959 |
-
"step": 122
|
| 960 |
-
},
|
| 961 |
-
{
|
| 962 |
-
"epoch": 3.4166666666666665,
|
| 963 |
-
"grad_norm": 0.6621596813201904,
|
| 964 |
-
"learning_rate": 6.771428571428572e-05,
|
| 965 |
-
"loss": 0.3273,
|
| 966 |
-
"step": 123
|
| 967 |
-
},
|
| 968 |
-
{
|
| 969 |
-
"epoch": 3.4444444444444446,
|
| 970 |
-
"grad_norm": 0.8255563974380493,
|
| 971 |
-
"learning_rate": 6.742857142857143e-05,
|
| 972 |
-
"loss": 0.3264,
|
| 973 |
-
"step": 124
|
| 974 |
-
},
|
| 975 |
-
{
|
| 976 |
-
"epoch": 3.4722222222222223,
|
| 977 |
-
"grad_norm": 0.8350001573562622,
|
| 978 |
-
"learning_rate": 6.714285714285714e-05,
|
| 979 |
-
"loss": 0.3157,
|
| 980 |
-
"step": 125
|
| 981 |
-
},
|
| 982 |
-
{
|
| 983 |
-
"epoch": 3.5,
|
| 984 |
-
"grad_norm": 0.7275986075401306,
|
| 985 |
-
"learning_rate": 6.685714285714286e-05,
|
| 986 |
-
"loss": 0.3205,
|
| 987 |
-
"step": 126
|
| 988 |
-
},
|
| 989 |
-
{
|
| 990 |
-
"epoch": 3.5277777777777777,
|
| 991 |
-
"grad_norm": 0.652642548084259,
|
| 992 |
-
"learning_rate": 6.657142857142857e-05,
|
| 993 |
-
"loss": 0.3327,
|
| 994 |
-
"step": 127
|
| 995 |
-
},
|
| 996 |
-
{
|
| 997 |
-
"epoch": 3.5555555555555554,
|
| 998 |
-
"grad_norm": 0.9522960186004639,
|
| 999 |
-
"learning_rate": 6.628571428571428e-05,
|
| 1000 |
-
"loss": 0.3319,
|
| 1001 |
-
"step": 128
|
| 1002 |
-
},
|
| 1003 |
-
{
|
| 1004 |
-
"epoch": 3.5833333333333335,
|
| 1005 |
-
"grad_norm": 0.7006963491439819,
|
| 1006 |
-
"learning_rate": 6.6e-05,
|
| 1007 |
-
"loss": 0.3292,
|
| 1008 |
-
"step": 129
|
| 1009 |
-
},
|
| 1010 |
-
{
|
| 1011 |
-
"epoch": 3.611111111111111,
|
| 1012 |
-
"grad_norm": 0.7161970138549805,
|
| 1013 |
-
"learning_rate": 6.571428571428571e-05,
|
| 1014 |
-
"loss": 0.3246,
|
| 1015 |
-
"step": 130
|
| 1016 |
-
},
|
| 1017 |
-
{
|
| 1018 |
-
"epoch": 3.611111111111111,
|
| 1019 |
-
"eval_loss": 0.3442412316799164,
|
| 1020 |
-
"eval_runtime": 0.1052,
|
| 1021 |
-
"eval_samples_per_second": 437.452,
|
| 1022 |
-
"eval_steps_per_second": 9.51,
|
| 1023 |
-
"step": 130
|
| 1024 |
-
},
|
| 1025 |
-
{
|
| 1026 |
-
"epoch": 3.638888888888889,
|
| 1027 |
-
"grad_norm": 1.0642709732055664,
|
| 1028 |
-
"learning_rate": 6.542857142857142e-05,
|
| 1029 |
-
"loss": 0.3289,
|
| 1030 |
-
"step": 131
|
| 1031 |
-
},
|
| 1032 |
-
{
|
| 1033 |
-
"epoch": 3.6666666666666665,
|
| 1034 |
-
"grad_norm": 0.7999193072319031,
|
| 1035 |
-
"learning_rate": 6.514285714285715e-05,
|
| 1036 |
-
"loss": 0.3234,
|
| 1037 |
-
"step": 132
|
| 1038 |
-
},
|
| 1039 |
-
{
|
| 1040 |
-
"epoch": 3.6944444444444446,
|
| 1041 |
-
"grad_norm": 0.8324876427650452,
|
| 1042 |
-
"learning_rate": 6.485714285714286e-05,
|
| 1043 |
-
"loss": 0.3297,
|
| 1044 |
-
"step": 133
|
| 1045 |
-
},
|
| 1046 |
-
{
|
| 1047 |
-
"epoch": 3.7222222222222223,
|
| 1048 |
-
"grad_norm": 0.561801552772522,
|
| 1049 |
-
"learning_rate": 6.457142857142856e-05,
|
| 1050 |
-
"loss": 0.3125,
|
| 1051 |
-
"step": 134
|
| 1052 |
-
},
|
| 1053 |
-
{
|
| 1054 |
-
"epoch": 3.75,
|
| 1055 |
-
"grad_norm": 0.6995918154716492,
|
| 1056 |
-
"learning_rate": 6.428571428571429e-05,
|
| 1057 |
-
"loss": 0.3234,
|
| 1058 |
-
"step": 135
|
| 1059 |
-
},
|
| 1060 |
-
{
|
| 1061 |
-
"epoch": 3.7777777777777777,
|
| 1062 |
-
"grad_norm": 0.6314477920532227,
|
| 1063 |
-
"learning_rate": 6.400000000000001e-05,
|
| 1064 |
-
"loss": 0.3256,
|
| 1065 |
-
"step": 136
|
| 1066 |
-
},
|
| 1067 |
-
{
|
| 1068 |
-
"epoch": 3.8055555555555554,
|
| 1069 |
-
"grad_norm": 0.9092559814453125,
|
| 1070 |
-
"learning_rate": 6.371428571428572e-05,
|
| 1071 |
-
"loss": 0.3315,
|
| 1072 |
-
"step": 137
|
| 1073 |
-
},
|
| 1074 |
-
{
|
| 1075 |
-
"epoch": 3.8333333333333335,
|
| 1076 |
-
"grad_norm": 0.7306588292121887,
|
| 1077 |
-
"learning_rate": 6.342857142857143e-05,
|
| 1078 |
-
"loss": 0.3241,
|
| 1079 |
-
"step": 138
|
| 1080 |
-
},
|
| 1081 |
-
{
|
| 1082 |
-
"epoch": 3.861111111111111,
|
| 1083 |
-
"grad_norm": 0.7943991422653198,
|
| 1084 |
-
"learning_rate": 6.314285714285715e-05,
|
| 1085 |
-
"loss": 0.3323,
|
| 1086 |
-
"step": 139
|
| 1087 |
-
},
|
| 1088 |
-
{
|
| 1089 |
-
"epoch": 3.888888888888889,
|
| 1090 |
-
"grad_norm": 0.8375313878059387,
|
| 1091 |
-
"learning_rate": 6.285714285714286e-05,
|
| 1092 |
-
"loss": 0.3273,
|
| 1093 |
-
"step": 140
|
| 1094 |
-
},
|
| 1095 |
-
{
|
| 1096 |
-
"epoch": 3.888888888888889,
|
| 1097 |
-
"eval_loss": 0.33936312794685364,
|
| 1098 |
-
"eval_runtime": 0.1048,
|
| 1099 |
-
"eval_samples_per_second": 438.851,
|
| 1100 |
-
"eval_steps_per_second": 9.54,
|
| 1101 |
-
"step": 140
|
| 1102 |
-
},
|
| 1103 |
-
{
|
| 1104 |
-
"epoch": 3.9166666666666665,
|
| 1105 |
-
"grad_norm": 0.9479944705963135,
|
| 1106 |
-
"learning_rate": 6.257142857142857e-05,
|
| 1107 |
-
"loss": 0.3258,
|
| 1108 |
-
"step": 141
|
| 1109 |
-
},
|
| 1110 |
-
{
|
| 1111 |
-
"epoch": 3.9444444444444446,
|
| 1112 |
-
"grad_norm": 0.8155922889709473,
|
| 1113 |
-
"learning_rate": 6.22857142857143e-05,
|
| 1114 |
-
"loss": 0.3228,
|
| 1115 |
-
"step": 142
|
| 1116 |
-
},
|
| 1117 |
-
{
|
| 1118 |
-
"epoch": 3.9722222222222223,
|
| 1119 |
-
"grad_norm": 0.8617050647735596,
|
| 1120 |
-
"learning_rate": 6.2e-05,
|
| 1121 |
-
"loss": 0.3217,
|
| 1122 |
-
"step": 143
|
| 1123 |
-
},
|
| 1124 |
-
{
|
| 1125 |
-
"epoch": 4.0,
|
| 1126 |
-
"grad_norm": 1.2106715440750122,
|
| 1127 |
-
"learning_rate": 6.171428571428571e-05,
|
| 1128 |
-
"loss": 0.3309,
|
| 1129 |
-
"step": 144
|
| 1130 |
-
},
|
| 1131 |
-
{
|
| 1132 |
-
"epoch": 4.027777777777778,
|
| 1133 |
-
"grad_norm": 0.8097350001335144,
|
| 1134 |
-
"learning_rate": 6.142857142857143e-05,
|
| 1135 |
-
"loss": 0.3265,
|
| 1136 |
-
"step": 145
|
| 1137 |
-
},
|
| 1138 |
-
{
|
| 1139 |
-
"epoch": 4.055555555555555,
|
| 1140 |
-
"grad_norm": 0.651019811630249,
|
| 1141 |
-
"learning_rate": 6.114285714285714e-05,
|
| 1142 |
-
"loss": 0.3222,
|
| 1143 |
-
"step": 146
|
| 1144 |
-
},
|
| 1145 |
-
{
|
| 1146 |
-
"epoch": 4.083333333333333,
|
| 1147 |
-
"grad_norm": 0.8858047127723694,
|
| 1148 |
-
"learning_rate": 6.085714285714286e-05,
|
| 1149 |
-
"loss": 0.3245,
|
| 1150 |
-
"step": 147
|
| 1151 |
-
},
|
| 1152 |
-
{
|
| 1153 |
-
"epoch": 4.111111111111111,
|
| 1154 |
-
"grad_norm": 0.8602396845817566,
|
| 1155 |
-
"learning_rate": 6.0571428571428576e-05,
|
| 1156 |
-
"loss": 0.3153,
|
| 1157 |
-
"step": 148
|
| 1158 |
-
},
|
| 1159 |
-
{
|
| 1160 |
-
"epoch": 4.138888888888889,
|
| 1161 |
-
"grad_norm": 0.615274965763092,
|
| 1162 |
-
"learning_rate": 6.028571428571429e-05,
|
| 1163 |
-
"loss": 0.313,
|
| 1164 |
-
"step": 149
|
| 1165 |
-
},
|
| 1166 |
-
{
|
| 1167 |
-
"epoch": 4.166666666666667,
|
| 1168 |
-
"grad_norm": 0.9199692010879517,
|
| 1169 |
-
"learning_rate": 6e-05,
|
| 1170 |
-
"loss": 0.3275,
|
| 1171 |
-
"step": 150
|
| 1172 |
-
},
|
| 1173 |
-
{
|
| 1174 |
-
"epoch": 4.166666666666667,
|
| 1175 |
-
"eval_loss": 0.34145644307136536,
|
| 1176 |
-
"eval_runtime": 0.1056,
|
| 1177 |
-
"eval_samples_per_second": 435.469,
|
| 1178 |
-
"eval_steps_per_second": 9.467,
|
| 1179 |
-
"step": 150
|
| 1180 |
-
},
|
| 1181 |
-
{
|
| 1182 |
-
"epoch": 4.194444444444445,
|
| 1183 |
-
"grad_norm": 0.9348899722099304,
|
| 1184 |
-
"learning_rate": 5.9714285714285724e-05,
|
| 1185 |
-
"loss": 0.3287,
|
| 1186 |
-
"step": 151
|
| 1187 |
-
},
|
| 1188 |
-
{
|
| 1189 |
-
"epoch": 4.222222222222222,
|
| 1190 |
-
"grad_norm": 0.7250774502754211,
|
| 1191 |
-
"learning_rate": 5.9428571428571434e-05,
|
| 1192 |
-
"loss": 0.3203,
|
| 1193 |
-
"step": 152
|
| 1194 |
-
},
|
| 1195 |
-
{
|
| 1196 |
-
"epoch": 4.25,
|
| 1197 |
-
"grad_norm": 0.7376280426979065,
|
| 1198 |
-
"learning_rate": 5.914285714285714e-05,
|
| 1199 |
-
"loss": 0.3169,
|
| 1200 |
-
"step": 153
|
| 1201 |
-
},
|
| 1202 |
-
{
|
| 1203 |
-
"epoch": 4.277777777777778,
|
| 1204 |
-
"grad_norm": 0.6010245680809021,
|
| 1205 |
-
"learning_rate": 5.885714285714285e-05,
|
| 1206 |
-
"loss": 0.3215,
|
| 1207 |
-
"step": 154
|
| 1208 |
-
},
|
| 1209 |
-
{
|
| 1210 |
-
"epoch": 4.305555555555555,
|
| 1211 |
-
"grad_norm": 0.7241640686988831,
|
| 1212 |
-
"learning_rate": 5.8571428571428575e-05,
|
| 1213 |
-
"loss": 0.317,
|
| 1214 |
-
"step": 155
|
| 1215 |
-
},
|
| 1216 |
-
{
|
| 1217 |
-
"epoch": 4.333333333333333,
|
| 1218 |
-
"grad_norm": 0.6956952810287476,
|
| 1219 |
-
"learning_rate": 5.828571428571429e-05,
|
| 1220 |
-
"loss": 0.3217,
|
| 1221 |
-
"step": 156
|
| 1222 |
-
},
|
| 1223 |
-
{
|
| 1224 |
-
"epoch": 4.361111111111111,
|
| 1225 |
-
"grad_norm": 0.8463672995567322,
|
| 1226 |
-
"learning_rate": 5.8e-05,
|
| 1227 |
-
"loss": 0.322,
|
| 1228 |
-
"step": 157
|
| 1229 |
-
},
|
| 1230 |
-
{
|
| 1231 |
-
"epoch": 4.388888888888889,
|
| 1232 |
-
"grad_norm": 0.5538536906242371,
|
| 1233 |
-
"learning_rate": 5.771428571428572e-05,
|
| 1234 |
-
"loss": 0.3129,
|
| 1235 |
-
"step": 158
|
| 1236 |
-
},
|
| 1237 |
-
{
|
| 1238 |
-
"epoch": 4.416666666666667,
|
| 1239 |
-
"grad_norm": 0.8398566246032715,
|
| 1240 |
-
"learning_rate": 5.742857142857143e-05,
|
| 1241 |
-
"loss": 0.3275,
|
| 1242 |
-
"step": 159
|
| 1243 |
-
},
|
| 1244 |
-
{
|
| 1245 |
-
"epoch": 4.444444444444445,
|
| 1246 |
-
"grad_norm": 0.5335714221000671,
|
| 1247 |
-
"learning_rate": 5.714285714285714e-05,
|
| 1248 |
-
"loss": 0.3225,
|
| 1249 |
-
"step": 160
|
| 1250 |
-
},
|
| 1251 |
-
{
|
| 1252 |
-
"epoch": 4.444444444444445,
|
| 1253 |
-
"eval_loss": 0.3419412672519684,
|
| 1254 |
-
"eval_runtime": 0.1054,
|
| 1255 |
-
"eval_samples_per_second": 436.373,
|
| 1256 |
-
"eval_steps_per_second": 9.486,
|
| 1257 |
-
"step": 160
|
| 1258 |
-
},
|
| 1259 |
-
{
|
| 1260 |
-
"epoch": 4.472222222222222,
|
| 1261 |
-
"grad_norm": 0.893516480922699,
|
| 1262 |
-
"learning_rate": 5.6857142857142865e-05,
|
| 1263 |
-
"loss": 0.3201,
|
| 1264 |
-
"step": 161
|
| 1265 |
-
},
|
| 1266 |
-
{
|
| 1267 |
-
"epoch": 4.5,
|
| 1268 |
-
"grad_norm": 0.7950851321220398,
|
| 1269 |
-
"learning_rate": 5.6571428571428574e-05,
|
| 1270 |
-
"loss": 0.3246,
|
| 1271 |
-
"step": 162
|
| 1272 |
-
},
|
| 1273 |
-
{
|
| 1274 |
-
"epoch": 4.527777777777778,
|
| 1275 |
-
"grad_norm": 0.6274649500846863,
|
| 1276 |
-
"learning_rate": 5.628571428571428e-05,
|
| 1277 |
-
"loss": 0.3138,
|
| 1278 |
-
"step": 163
|
| 1279 |
-
},
|
| 1280 |
-
{
|
| 1281 |
-
"epoch": 4.555555555555555,
|
| 1282 |
-
"grad_norm": 0.6270793676376343,
|
| 1283 |
-
"learning_rate": 5.6000000000000006e-05,
|
| 1284 |
-
"loss": 0.3225,
|
| 1285 |
-
"step": 164
|
| 1286 |
-
},
|
| 1287 |
-
{
|
| 1288 |
-
"epoch": 4.583333333333333,
|
| 1289 |
-
"grad_norm": 0.6661616563796997,
|
| 1290 |
-
"learning_rate": 5.571428571428572e-05,
|
| 1291 |
-
"loss": 0.3222,
|
| 1292 |
-
"step": 165
|
| 1293 |
-
},
|
| 1294 |
-
{
|
| 1295 |
-
"epoch": 4.611111111111111,
|
| 1296 |
-
"grad_norm": 0.6097862124443054,
|
| 1297 |
-
"learning_rate": 5.542857142857143e-05,
|
| 1298 |
-
"loss": 0.3138,
|
| 1299 |
-
"step": 166
|
| 1300 |
-
},
|
| 1301 |
-
{
|
| 1302 |
-
"epoch": 4.638888888888889,
|
| 1303 |
-
"grad_norm": 0.6743194460868835,
|
| 1304 |
-
"learning_rate": 5.514285714285714e-05,
|
| 1305 |
-
"loss": 0.3109,
|
| 1306 |
-
"step": 167
|
| 1307 |
-
},
|
| 1308 |
-
{
|
| 1309 |
-
"epoch": 4.666666666666667,
|
| 1310 |
-
"grad_norm": 0.6684880256652832,
|
| 1311 |
-
"learning_rate": 5.485714285714286e-05,
|
| 1312 |
-
"loss": 0.3193,
|
| 1313 |
-
"step": 168
|
| 1314 |
-
},
|
| 1315 |
-
{
|
| 1316 |
-
"epoch": 4.694444444444445,
|
| 1317 |
-
"grad_norm": 0.7434603571891785,
|
| 1318 |
-
"learning_rate": 5.457142857142857e-05,
|
| 1319 |
-
"loss": 0.3212,
|
| 1320 |
-
"step": 169
|
| 1321 |
-
},
|
| 1322 |
-
{
|
| 1323 |
-
"epoch": 4.722222222222222,
|
| 1324 |
-
"grad_norm": 0.8257206082344055,
|
| 1325 |
-
"learning_rate": 5.428571428571428e-05,
|
| 1326 |
-
"loss": 0.323,
|
| 1327 |
-
"step": 170
|
| 1328 |
-
},
|
| 1329 |
-
{
|
| 1330 |
-
"epoch": 4.722222222222222,
|
| 1331 |
-
"eval_loss": 0.335285484790802,
|
| 1332 |
-
"eval_runtime": 0.1056,
|
| 1333 |
-
"eval_samples_per_second": 435.635,
|
| 1334 |
-
"eval_steps_per_second": 9.47,
|
| 1335 |
-
"step": 170
|
| 1336 |
-
},
|
| 1337 |
-
{
|
| 1338 |
-
"epoch": 4.75,
|
| 1339 |
-
"grad_norm": 0.5611714720726013,
|
| 1340 |
-
"learning_rate": 5.4000000000000005e-05,
|
| 1341 |
-
"loss": 0.3249,
|
| 1342 |
-
"step": 171
|
| 1343 |
-
},
|
| 1344 |
-
{
|
| 1345 |
-
"epoch": 4.777777777777778,
|
| 1346 |
-
"grad_norm": 0.6146838068962097,
|
| 1347 |
-
"learning_rate": 5.3714285714285714e-05,
|
| 1348 |
-
"loss": 0.3213,
|
| 1349 |
-
"step": 172
|
| 1350 |
-
},
|
| 1351 |
-
{
|
| 1352 |
-
"epoch": 4.805555555555555,
|
| 1353 |
-
"grad_norm": 0.6939036846160889,
|
| 1354 |
-
"learning_rate": 5.342857142857143e-05,
|
| 1355 |
-
"loss": 0.3239,
|
| 1356 |
-
"step": 173
|
| 1357 |
-
},
|
| 1358 |
-
{
|
| 1359 |
-
"epoch": 4.833333333333333,
|
| 1360 |
-
"grad_norm": 0.7213876843452454,
|
| 1361 |
-
"learning_rate": 5.314285714285715e-05,
|
| 1362 |
-
"loss": 0.3234,
|
| 1363 |
-
"step": 174
|
| 1364 |
-
},
|
| 1365 |
-
{
|
| 1366 |
-
"epoch": 4.861111111111111,
|
| 1367 |
-
"grad_norm": 0.6637408137321472,
|
| 1368 |
-
"learning_rate": 5.285714285714286e-05,
|
| 1369 |
-
"loss": 0.3186,
|
| 1370 |
-
"step": 175
|
| 1371 |
-
},
|
| 1372 |
-
{
|
| 1373 |
-
"epoch": 4.888888888888889,
|
| 1374 |
-
"grad_norm": 0.6469508409500122,
|
| 1375 |
-
"learning_rate": 5.257142857142857e-05,
|
| 1376 |
-
"loss": 0.3184,
|
| 1377 |
-
"step": 176
|
| 1378 |
-
},
|
| 1379 |
-
{
|
| 1380 |
-
"epoch": 4.916666666666667,
|
| 1381 |
-
"grad_norm": 0.6262702941894531,
|
| 1382 |
-
"learning_rate": 5.2285714285714294e-05,
|
| 1383 |
-
"loss": 0.3171,
|
| 1384 |
-
"step": 177
|
| 1385 |
-
},
|
| 1386 |
-
{
|
| 1387 |
-
"epoch": 4.944444444444445,
|
| 1388 |
-
"grad_norm": 0.6692309379577637,
|
| 1389 |
-
"learning_rate": 5.2000000000000004e-05,
|
| 1390 |
-
"loss": 0.3271,
|
| 1391 |
-
"step": 178
|
| 1392 |
-
},
|
| 1393 |
-
{
|
| 1394 |
-
"epoch": 4.972222222222222,
|
| 1395 |
-
"grad_norm": 0.611004114151001,
|
| 1396 |
-
"learning_rate": 5.171428571428571e-05,
|
| 1397 |
-
"loss": 0.3229,
|
| 1398 |
-
"step": 179
|
| 1399 |
-
},
|
| 1400 |
-
{
|
| 1401 |
-
"epoch": 5.0,
|
| 1402 |
-
"grad_norm": 0.9707463383674622,
|
| 1403 |
-
"learning_rate": 5.142857142857143e-05,
|
| 1404 |
-
"loss": 0.3223,
|
| 1405 |
-
"step": 180
|
| 1406 |
-
},
|
| 1407 |
-
{
|
| 1408 |
-
"epoch": 5.0,
|
| 1409 |
-
"eval_loss": 0.33622071146965027,
|
| 1410 |
-
"eval_runtime": 0.105,
|
| 1411 |
-
"eval_samples_per_second": 437.997,
|
| 1412 |
-
"eval_steps_per_second": 9.522,
|
| 1413 |
-
"step": 180
|
| 1414 |
-
},
|
| 1415 |
-
{
|
| 1416 |
-
"epoch": 5.027777777777778,
|
| 1417 |
-
"grad_norm": 0.43101295828819275,
|
| 1418 |
-
"learning_rate": 5.1142857142857145e-05,
|
| 1419 |
-
"loss": 0.3106,
|
| 1420 |
-
"step": 181
|
| 1421 |
-
},
|
| 1422 |
-
{
|
| 1423 |
-
"epoch": 5.055555555555555,
|
| 1424 |
-
"grad_norm": 0.7981957793235779,
|
| 1425 |
-
"learning_rate": 5.085714285714286e-05,
|
| 1426 |
-
"loss": 0.3139,
|
| 1427 |
-
"step": 182
|
| 1428 |
-
},
|
| 1429 |
-
{
|
| 1430 |
-
"epoch": 5.083333333333333,
|
| 1431 |
-
"grad_norm": 0.9149967432022095,
|
| 1432 |
-
"learning_rate": 5.057142857142857e-05,
|
| 1433 |
-
"loss": 0.3137,
|
| 1434 |
-
"step": 183
|
| 1435 |
-
},
|
| 1436 |
-
{
|
| 1437 |
-
"epoch": 5.111111111111111,
|
| 1438 |
-
"grad_norm": 0.8689376711845398,
|
| 1439 |
-
"learning_rate": 5.028571428571429e-05,
|
| 1440 |
-
"loss": 0.3185,
|
| 1441 |
-
"step": 184
|
| 1442 |
-
},
|
| 1443 |
-
{
|
| 1444 |
-
"epoch": 5.138888888888889,
|
| 1445 |
-
"grad_norm": 0.6829914450645447,
|
| 1446 |
-
"learning_rate": 5e-05,
|
| 1447 |
-
"loss": 0.3195,
|
| 1448 |
-
"step": 185
|
| 1449 |
-
},
|
| 1450 |
-
{
|
| 1451 |
-
"epoch": 5.166666666666667,
|
| 1452 |
-
"grad_norm": 0.6187098026275635,
|
| 1453 |
-
"learning_rate": 4.971428571428572e-05,
|
| 1454 |
-
"loss": 0.3139,
|
| 1455 |
-
"step": 186
|
| 1456 |
-
},
|
| 1457 |
-
{
|
| 1458 |
-
"epoch": 5.194444444444445,
|
| 1459 |
-
"grad_norm": 0.8703141212463379,
|
| 1460 |
-
"learning_rate": 4.942857142857143e-05,
|
| 1461 |
-
"loss": 0.3147,
|
| 1462 |
-
"step": 187
|
| 1463 |
-
},
|
| 1464 |
-
{
|
| 1465 |
-
"epoch": 5.222222222222222,
|
| 1466 |
-
"grad_norm": 0.6344360709190369,
|
| 1467 |
-
"learning_rate": 4.9142857142857144e-05,
|
| 1468 |
-
"loss": 0.3162,
|
| 1469 |
-
"step": 188
|
| 1470 |
-
},
|
| 1471 |
-
{
|
| 1472 |
-
"epoch": 5.25,
|
| 1473 |
-
"grad_norm": 0.7499691843986511,
|
| 1474 |
-
"learning_rate": 4.885714285714286e-05,
|
| 1475 |
-
"loss": 0.3228,
|
| 1476 |
-
"step": 189
|
| 1477 |
-
},
|
| 1478 |
-
{
|
| 1479 |
-
"epoch": 5.277777777777778,
|
| 1480 |
-
"grad_norm": 0.7664843201637268,
|
| 1481 |
-
"learning_rate": 4.8571428571428576e-05,
|
| 1482 |
-
"loss": 0.3152,
|
| 1483 |
-
"step": 190
|
| 1484 |
-
},
|
| 1485 |
-
{
|
| 1486 |
-
"epoch": 5.277777777777778,
|
| 1487 |
-
"eval_loss": 0.3364305794239044,
|
| 1488 |
-
"eval_runtime": 0.1056,
|
| 1489 |
-
"eval_samples_per_second": 435.778,
|
| 1490 |
-
"eval_steps_per_second": 9.473,
|
| 1491 |
-
"step": 190
|
| 1492 |
-
},
|
| 1493 |
-
{
|
| 1494 |
-
"epoch": 5.305555555555555,
|
| 1495 |
-
"grad_norm": 0.6158504486083984,
|
| 1496 |
-
"learning_rate": 4.828571428571429e-05,
|
| 1497 |
-
"loss": 0.3154,
|
| 1498 |
-
"step": 191
|
| 1499 |
-
},
|
| 1500 |
-
{
|
| 1501 |
-
"epoch": 5.333333333333333,
|
| 1502 |
-
"grad_norm": 0.8614490032196045,
|
| 1503 |
-
"learning_rate": 4.8e-05,
|
| 1504 |
-
"loss": 0.3206,
|
| 1505 |
-
"step": 192
|
| 1506 |
-
},
|
| 1507 |
-
{
|
| 1508 |
-
"epoch": 5.361111111111111,
|
| 1509 |
-
"grad_norm": 0.7699540257453918,
|
| 1510 |
-
"learning_rate": 4.771428571428572e-05,
|
| 1511 |
-
"loss": 0.3159,
|
| 1512 |
-
"step": 193
|
| 1513 |
-
},
|
| 1514 |
-
{
|
| 1515 |
-
"epoch": 5.388888888888889,
|
| 1516 |
-
"grad_norm": 0.9598901867866516,
|
| 1517 |
-
"learning_rate": 4.742857142857143e-05,
|
| 1518 |
-
"loss": 0.3186,
|
| 1519 |
-
"step": 194
|
| 1520 |
-
},
|
| 1521 |
-
{
|
| 1522 |
-
"epoch": 5.416666666666667,
|
| 1523 |
-
"grad_norm": 0.855253279209137,
|
| 1524 |
-
"learning_rate": 4.714285714285714e-05,
|
| 1525 |
-
"loss": 0.3118,
|
| 1526 |
-
"step": 195
|
| 1527 |
-
},
|
| 1528 |
-
{
|
| 1529 |
-
"epoch": 5.444444444444445,
|
| 1530 |
-
"grad_norm": 0.6478847861289978,
|
| 1531 |
-
"learning_rate": 4.685714285714286e-05,
|
| 1532 |
-
"loss": 0.3178,
|
| 1533 |
-
"step": 196
|
| 1534 |
-
},
|
| 1535 |
-
{
|
| 1536 |
-
"epoch": 5.472222222222222,
|
| 1537 |
-
"grad_norm": 0.8028067946434021,
|
| 1538 |
-
"learning_rate": 4.6571428571428575e-05,
|
| 1539 |
-
"loss": 0.3236,
|
| 1540 |
-
"step": 197
|
| 1541 |
-
},
|
| 1542 |
-
{
|
| 1543 |
-
"epoch": 5.5,
|
| 1544 |
-
"grad_norm": 0.7795782089233398,
|
| 1545 |
-
"learning_rate": 4.628571428571429e-05,
|
| 1546 |
-
"loss": 0.3147,
|
| 1547 |
-
"step": 198
|
| 1548 |
-
},
|
| 1549 |
-
{
|
| 1550 |
-
"epoch": 5.527777777777778,
|
| 1551 |
-
"grad_norm": 0.7845653891563416,
|
| 1552 |
-
"learning_rate": 4.600000000000001e-05,
|
| 1553 |
-
"loss": 0.3221,
|
| 1554 |
-
"step": 199
|
| 1555 |
-
},
|
| 1556 |
-
{
|
| 1557 |
-
"epoch": 5.555555555555555,
|
| 1558 |
-
"grad_norm": 1.1422370672225952,
|
| 1559 |
-
"learning_rate": 4.5714285714285716e-05,
|
| 1560 |
-
"loss": 0.321,
|
| 1561 |
-
"step": 200
|
| 1562 |
-
},
|
| 1563 |
-
{
|
| 1564 |
-
"epoch": 5.555555555555555,
|
| 1565 |
-
"eval_loss": 0.33535492420196533,
|
| 1566 |
-
"eval_runtime": 0.1051,
|
| 1567 |
-
"eval_samples_per_second": 437.57,
|
| 1568 |
-
"eval_steps_per_second": 9.512,
|
| 1569 |
-
"step": 200
|
| 1570 |
-
},
|
| 1571 |
-
{
|
| 1572 |
-
"epoch": 5.583333333333333,
|
| 1573 |
-
"grad_norm": 0.7386415600776672,
|
| 1574 |
-
"learning_rate": 4.542857142857143e-05,
|
| 1575 |
-
"loss": 0.3183,
|
| 1576 |
-
"step": 201
|
| 1577 |
-
},
|
| 1578 |
-
{
|
| 1579 |
-
"epoch": 5.611111111111111,
|
| 1580 |
-
"grad_norm": 0.6756716966629028,
|
| 1581 |
-
"learning_rate": 4.514285714285714e-05,
|
| 1582 |
-
"loss": 0.3205,
|
| 1583 |
-
"step": 202
|
| 1584 |
-
},
|
| 1585 |
-
{
|
| 1586 |
-
"epoch": 5.638888888888889,
|
| 1587 |
-
"grad_norm": 0.7116839289665222,
|
| 1588 |
-
"learning_rate": 4.485714285714286e-05,
|
| 1589 |
-
"loss": 0.3195,
|
| 1590 |
-
"step": 203
|
| 1591 |
-
},
|
| 1592 |
-
{
|
| 1593 |
-
"epoch": 5.666666666666667,
|
| 1594 |
-
"grad_norm": 0.7919530272483826,
|
| 1595 |
-
"learning_rate": 4.4571428571428574e-05,
|
| 1596 |
-
"loss": 0.3248,
|
| 1597 |
-
"step": 204
|
| 1598 |
-
},
|
| 1599 |
-
{
|
| 1600 |
-
"epoch": 5.694444444444445,
|
| 1601 |
-
"grad_norm": 0.5152342319488525,
|
| 1602 |
-
"learning_rate": 4.428571428571428e-05,
|
| 1603 |
-
"loss": 0.3124,
|
| 1604 |
-
"step": 205
|
| 1605 |
-
},
|
| 1606 |
-
{
|
| 1607 |
-
"epoch": 5.722222222222222,
|
| 1608 |
-
"grad_norm": 0.9519732594490051,
|
| 1609 |
-
"learning_rate": 4.4000000000000006e-05,
|
| 1610 |
-
"loss": 0.3205,
|
| 1611 |
-
"step": 206
|
| 1612 |
-
},
|
| 1613 |
-
{
|
| 1614 |
-
"epoch": 5.75,
|
| 1615 |
-
"grad_norm": 0.7759018540382385,
|
| 1616 |
-
"learning_rate": 4.371428571428572e-05,
|
| 1617 |
-
"loss": 0.3188,
|
| 1618 |
-
"step": 207
|
| 1619 |
-
},
|
| 1620 |
-
{
|
| 1621 |
-
"epoch": 5.777777777777778,
|
| 1622 |
-
"grad_norm": 0.931468665599823,
|
| 1623 |
-
"learning_rate": 4.342857142857143e-05,
|
| 1624 |
-
"loss": 0.3195,
|
| 1625 |
-
"step": 208
|
| 1626 |
-
},
|
| 1627 |
-
{
|
| 1628 |
-
"epoch": 5.805555555555555,
|
| 1629 |
-
"grad_norm": 1.0431036949157715,
|
| 1630 |
-
"learning_rate": 4.314285714285715e-05,
|
| 1631 |
-
"loss": 0.3256,
|
| 1632 |
-
"step": 209
|
| 1633 |
-
},
|
| 1634 |
-
{
|
| 1635 |
-
"epoch": 5.833333333333333,
|
| 1636 |
-
"grad_norm": 0.6952974200248718,
|
| 1637 |
-
"learning_rate": 4.2857142857142856e-05,
|
| 1638 |
-
"loss": 0.3189,
|
| 1639 |
-
"step": 210
|
| 1640 |
-
},
|
| 1641 |
-
{
|
| 1642 |
-
"epoch": 5.833333333333333,
|
| 1643 |
-
"eval_loss": 0.33737656474113464,
|
| 1644 |
-
"eval_runtime": 0.1056,
|
| 1645 |
-
"eval_samples_per_second": 435.436,
|
| 1646 |
-
"eval_steps_per_second": 9.466,
|
| 1647 |
-
"step": 210
|
| 1648 |
-
},
|
| 1649 |
-
{
|
| 1650 |
-
"epoch": 5.861111111111111,
|
| 1651 |
-
"grad_norm": 0.880944013595581,
|
| 1652 |
-
"learning_rate": 4.257142857142857e-05,
|
| 1653 |
-
"loss": 0.3159,
|
| 1654 |
-
"step": 211
|
| 1655 |
-
},
|
| 1656 |
-
{
|
| 1657 |
-
"epoch": 5.888888888888889,
|
| 1658 |
-
"grad_norm": 0.6606886982917786,
|
| 1659 |
-
"learning_rate": 4.228571428571429e-05,
|
| 1660 |
-
"loss": 0.3177,
|
| 1661 |
-
"step": 212
|
| 1662 |
-
},
|
| 1663 |
-
{
|
| 1664 |
-
"epoch": 5.916666666666667,
|
| 1665 |
-
"grad_norm": 0.7617785930633545,
|
| 1666 |
-
"learning_rate": 4.2e-05,
|
| 1667 |
-
"loss": 0.317,
|
| 1668 |
-
"step": 213
|
| 1669 |
-
},
|
| 1670 |
-
{
|
| 1671 |
-
"epoch": 5.944444444444445,
|
| 1672 |
-
"grad_norm": 0.6381199955940247,
|
| 1673 |
-
"learning_rate": 4.1714285714285714e-05,
|
| 1674 |
-
"loss": 0.313,
|
| 1675 |
-
"step": 214
|
| 1676 |
-
},
|
| 1677 |
-
{
|
| 1678 |
-
"epoch": 5.972222222222222,
|
| 1679 |
-
"grad_norm": 0.6644593477249146,
|
| 1680 |
-
"learning_rate": 4.1428571428571437e-05,
|
| 1681 |
-
"loss": 0.3184,
|
| 1682 |
-
"step": 215
|
| 1683 |
-
},
|
| 1684 |
-
{
|
| 1685 |
-
"epoch": 6.0,
|
| 1686 |
-
"grad_norm": 0.9742204546928406,
|
| 1687 |
-
"learning_rate": 4.1142857142857146e-05,
|
| 1688 |
-
"loss": 0.3103,
|
| 1689 |
-
"step": 216
|
| 1690 |
-
},
|
| 1691 |
-
{
|
| 1692 |
-
"epoch": 6.027777777777778,
|
| 1693 |
-
"grad_norm": 0.6401204466819763,
|
| 1694 |
-
"learning_rate": 4.085714285714286e-05,
|
| 1695 |
-
"loss": 0.311,
|
| 1696 |
-
"step": 217
|
| 1697 |
-
},
|
| 1698 |
-
{
|
| 1699 |
-
"epoch": 6.055555555555555,
|
| 1700 |
-
"grad_norm": 0.785503089427948,
|
| 1701 |
-
"learning_rate": 4.057142857142857e-05,
|
| 1702 |
-
"loss": 0.3155,
|
| 1703 |
-
"step": 218
|
| 1704 |
-
},
|
| 1705 |
-
{
|
| 1706 |
-
"epoch": 6.083333333333333,
|
| 1707 |
-
"grad_norm": 0.5417785048484802,
|
| 1708 |
-
"learning_rate": 4.028571428571429e-05,
|
| 1709 |
-
"loss": 0.3109,
|
| 1710 |
-
"step": 219
|
| 1711 |
-
},
|
| 1712 |
-
{
|
| 1713 |
-
"epoch": 6.111111111111111,
|
| 1714 |
-
"grad_norm": 0.6186631321907043,
|
| 1715 |
-
"learning_rate": 4e-05,
|
| 1716 |
-
"loss": 0.3114,
|
| 1717 |
-
"step": 220
|
| 1718 |
-
},
|
| 1719 |
-
{
|
| 1720 |
-
"epoch": 6.111111111111111,
|
| 1721 |
-
"eval_loss": 0.32762280106544495,
|
| 1722 |
-
"eval_runtime": 0.1058,
|
| 1723 |
-
"eval_samples_per_second": 434.862,
|
| 1724 |
-
"eval_steps_per_second": 9.454,
|
| 1725 |
-
"step": 220
|
| 1726 |
-
},
|
| 1727 |
-
{
|
| 1728 |
-
"epoch": 6.138888888888889,
|
| 1729 |
-
"grad_norm": 0.6042884588241577,
|
| 1730 |
-
"learning_rate": 3.971428571428571e-05,
|
| 1731 |
-
"loss": 0.3099,
|
| 1732 |
-
"step": 221
|
| 1733 |
-
},
|
| 1734 |
-
{
|
| 1735 |
-
"epoch": 6.166666666666667,
|
| 1736 |
-
"grad_norm": 0.5823580026626587,
|
| 1737 |
-
"learning_rate": 3.942857142857143e-05,
|
| 1738 |
-
"loss": 0.3104,
|
| 1739 |
-
"step": 222
|
| 1740 |
-
},
|
| 1741 |
-
{
|
| 1742 |
-
"epoch": 6.194444444444445,
|
| 1743 |
-
"grad_norm": 0.6572258472442627,
|
| 1744 |
-
"learning_rate": 3.9142857142857145e-05,
|
| 1745 |
-
"loss": 0.3154,
|
| 1746 |
-
"step": 223
|
| 1747 |
-
},
|
| 1748 |
-
{
|
| 1749 |
-
"epoch": 6.222222222222222,
|
| 1750 |
-
"grad_norm": 0.565834641456604,
|
| 1751 |
-
"learning_rate": 3.885714285714286e-05,
|
| 1752 |
-
"loss": 0.3121,
|
| 1753 |
-
"step": 224
|
| 1754 |
-
},
|
| 1755 |
-
{
|
| 1756 |
-
"epoch": 6.25,
|
| 1757 |
-
"grad_norm": 0.7184673547744751,
|
| 1758 |
-
"learning_rate": 3.857142857142858e-05,
|
| 1759 |
-
"loss": 0.3119,
|
| 1760 |
-
"step": 225
|
| 1761 |
-
},
|
| 1762 |
-
{
|
| 1763 |
-
"epoch": 6.277777777777778,
|
| 1764 |
-
"grad_norm": 0.7170347571372986,
|
| 1765 |
-
"learning_rate": 3.8285714285714286e-05,
|
| 1766 |
-
"loss": 0.3158,
|
| 1767 |
-
"step": 226
|
| 1768 |
-
},
|
| 1769 |
-
{
|
| 1770 |
-
"epoch": 6.305555555555555,
|
| 1771 |
-
"grad_norm": 0.6102560758590698,
|
| 1772 |
-
"learning_rate": 3.8e-05,
|
| 1773 |
-
"loss": 0.3038,
|
| 1774 |
-
"step": 227
|
| 1775 |
-
},
|
| 1776 |
-
{
|
| 1777 |
-
"epoch": 6.333333333333333,
|
| 1778 |
-
"grad_norm": 0.7612823843955994,
|
| 1779 |
-
"learning_rate": 3.771428571428572e-05,
|
| 1780 |
-
"loss": 0.3124,
|
| 1781 |
-
"step": 228
|
| 1782 |
-
},
|
| 1783 |
-
{
|
| 1784 |
-
"epoch": 6.361111111111111,
|
| 1785 |
-
"grad_norm": 0.6277872920036316,
|
| 1786 |
-
"learning_rate": 3.742857142857143e-05,
|
| 1787 |
-
"loss": 0.3028,
|
| 1788 |
-
"step": 229
|
| 1789 |
-
},
|
| 1790 |
-
{
|
| 1791 |
-
"epoch": 6.388888888888889,
|
| 1792 |
-
"grad_norm": 0.7007192373275757,
|
| 1793 |
-
"learning_rate": 3.7142857142857143e-05,
|
| 1794 |
-
"loss": 0.3202,
|
| 1795 |
-
"step": 230
|
| 1796 |
-
},
|
| 1797 |
-
{
|
| 1798 |
-
"epoch": 6.388888888888889,
|
| 1799 |
-
"eval_loss": 0.33139991760253906,
|
| 1800 |
-
"eval_runtime": 0.1062,
|
| 1801 |
-
"eval_samples_per_second": 433.311,
|
| 1802 |
-
"eval_steps_per_second": 9.42,
|
| 1803 |
-
"step": 230
|
| 1804 |
-
},
|
| 1805 |
-
{
|
| 1806 |
-
"epoch": 6.416666666666667,
|
| 1807 |
-
"grad_norm": 0.6396629810333252,
|
| 1808 |
-
"learning_rate": 3.685714285714286e-05,
|
| 1809 |
-
"loss": 0.313,
|
| 1810 |
-
"step": 231
|
| 1811 |
-
},
|
| 1812 |
-
{
|
| 1813 |
-
"epoch": 6.444444444444445,
|
| 1814 |
-
"grad_norm": 0.5031012892723083,
|
| 1815 |
-
"learning_rate": 3.6571428571428576e-05,
|
| 1816 |
-
"loss": 0.3116,
|
| 1817 |
-
"step": 232
|
| 1818 |
-
},
|
| 1819 |
-
{
|
| 1820 |
-
"epoch": 6.472222222222222,
|
| 1821 |
-
"grad_norm": 0.7323219776153564,
|
| 1822 |
-
"learning_rate": 3.628571428571429e-05,
|
| 1823 |
-
"loss": 0.3172,
|
| 1824 |
-
"step": 233
|
| 1825 |
-
},
|
| 1826 |
-
{
|
| 1827 |
-
"epoch": 6.5,
|
| 1828 |
-
"grad_norm": 0.9094661474227905,
|
| 1829 |
-
"learning_rate": 3.6e-05,
|
| 1830 |
-
"loss": 0.3103,
|
| 1831 |
-
"step": 234
|
| 1832 |
-
},
|
| 1833 |
-
{
|
| 1834 |
-
"epoch": 6.527777777777778,
|
| 1835 |
-
"grad_norm": 0.5560885667800903,
|
| 1836 |
-
"learning_rate": 3.571428571428572e-05,
|
| 1837 |
-
"loss": 0.3056,
|
| 1838 |
-
"step": 235
|
| 1839 |
-
},
|
| 1840 |
-
{
|
| 1841 |
-
"epoch": 6.555555555555555,
|
| 1842 |
-
"grad_norm": 1.0145907402038574,
|
| 1843 |
-
"learning_rate": 3.5428571428571426e-05,
|
| 1844 |
-
"loss": 0.3096,
|
| 1845 |
-
"step": 236
|
| 1846 |
-
},
|
| 1847 |
-
{
|
| 1848 |
-
"epoch": 6.583333333333333,
|
| 1849 |
-
"grad_norm": 0.8287002444267273,
|
| 1850 |
-
"learning_rate": 3.514285714285714e-05,
|
| 1851 |
-
"loss": 0.3049,
|
| 1852 |
-
"step": 237
|
| 1853 |
-
},
|
| 1854 |
-
{
|
| 1855 |
-
"epoch": 6.611111111111111,
|
| 1856 |
-
"grad_norm": 0.5207920074462891,
|
| 1857 |
-
"learning_rate": 3.485714285714286e-05,
|
| 1858 |
-
"loss": 0.3047,
|
| 1859 |
-
"step": 238
|
| 1860 |
-
},
|
| 1861 |
-
{
|
| 1862 |
-
"epoch": 6.638888888888889,
|
| 1863 |
-
"grad_norm": 1.065272331237793,
|
| 1864 |
-
"learning_rate": 3.4571428571428574e-05,
|
| 1865 |
-
"loss": 0.3156,
|
| 1866 |
-
"step": 239
|
| 1867 |
-
},
|
| 1868 |
-
{
|
| 1869 |
-
"epoch": 6.666666666666667,
|
| 1870 |
-
"grad_norm": 0.6712301969528198,
|
| 1871 |
-
"learning_rate": 3.428571428571429e-05,
|
| 1872 |
-
"loss": 0.3055,
|
| 1873 |
-
"step": 240
|
| 1874 |
-
},
|
| 1875 |
-
{
|
| 1876 |
-
"epoch": 6.666666666666667,
|
| 1877 |
-
"eval_loss": 0.3387901484966278,
|
| 1878 |
-
"eval_runtime": 0.1053,
|
| 1879 |
-
"eval_samples_per_second": 436.76,
|
| 1880 |
-
"eval_steps_per_second": 9.495,
|
| 1881 |
-
"step": 240
|
| 1882 |
-
},
|
| 1883 |
-
{
|
| 1884 |
-
"epoch": 6.694444444444445,
|
| 1885 |
-
"grad_norm": 0.9477025866508484,
|
| 1886 |
-
"learning_rate": 3.4000000000000007e-05,
|
| 1887 |
-
"loss": 0.3164,
|
| 1888 |
-
"step": 241
|
| 1889 |
-
},
|
| 1890 |
-
{
|
| 1891 |
-
"epoch": 6.722222222222222,
|
| 1892 |
-
"grad_norm": 1.179295301437378,
|
| 1893 |
-
"learning_rate": 3.3714285714285716e-05,
|
| 1894 |
-
"loss": 0.3263,
|
| 1895 |
-
"step": 242
|
| 1896 |
-
},
|
| 1897 |
-
{
|
| 1898 |
-
"epoch": 6.75,
|
| 1899 |
-
"grad_norm": 0.6969395875930786,
|
| 1900 |
-
"learning_rate": 3.342857142857143e-05,
|
| 1901 |
-
"loss": 0.3136,
|
| 1902 |
-
"step": 243
|
| 1903 |
-
},
|
| 1904 |
-
{
|
| 1905 |
-
"epoch": 6.777777777777778,
|
| 1906 |
-
"grad_norm": 0.7752478718757629,
|
| 1907 |
-
"learning_rate": 3.314285714285714e-05,
|
| 1908 |
-
"loss": 0.3143,
|
| 1909 |
-
"step": 244
|
| 1910 |
-
},
|
| 1911 |
-
{
|
| 1912 |
-
"epoch": 6.805555555555555,
|
| 1913 |
-
"grad_norm": 0.7640174031257629,
|
| 1914 |
-
"learning_rate": 3.285714285714286e-05,
|
| 1915 |
-
"loss": 0.3143,
|
| 1916 |
-
"step": 245
|
| 1917 |
-
},
|
| 1918 |
-
{
|
| 1919 |
-
"epoch": 6.833333333333333,
|
| 1920 |
-
"grad_norm": 0.9904161691665649,
|
| 1921 |
-
"learning_rate": 3.257142857142857e-05,
|
| 1922 |
-
"loss": 0.3093,
|
| 1923 |
-
"step": 246
|
| 1924 |
-
},
|
| 1925 |
-
{
|
| 1926 |
-
"epoch": 6.861111111111111,
|
| 1927 |
-
"grad_norm": 0.7707158923149109,
|
| 1928 |
-
"learning_rate": 3.228571428571428e-05,
|
| 1929 |
-
"loss": 0.3185,
|
| 1930 |
-
"step": 247
|
| 1931 |
-
},
|
| 1932 |
-
{
|
| 1933 |
-
"epoch": 6.888888888888889,
|
| 1934 |
-
"grad_norm": 0.8660508394241333,
|
| 1935 |
-
"learning_rate": 3.2000000000000005e-05,
|
| 1936 |
-
"loss": 0.3122,
|
| 1937 |
-
"step": 248
|
| 1938 |
-
},
|
| 1939 |
-
{
|
| 1940 |
-
"epoch": 6.916666666666667,
|
| 1941 |
-
"grad_norm": 0.7438889741897583,
|
| 1942 |
-
"learning_rate": 3.1714285714285715e-05,
|
| 1943 |
-
"loss": 0.3148,
|
| 1944 |
-
"step": 249
|
| 1945 |
-
},
|
| 1946 |
-
{
|
| 1947 |
-
"epoch": 6.944444444444445,
|
| 1948 |
-
"grad_norm": 0.5746331810951233,
|
| 1949 |
-
"learning_rate": 3.142857142857143e-05,
|
| 1950 |
-
"loss": 0.3137,
|
| 1951 |
-
"step": 250
|
| 1952 |
-
},
|
| 1953 |
-
{
|
| 1954 |
-
"epoch": 6.944444444444445,
|
| 1955 |
-
"eval_loss": 0.32808226346969604,
|
| 1956 |
-
"eval_runtime": 0.1059,
|
| 1957 |
-
"eval_samples_per_second": 434.438,
|
| 1958 |
-
"eval_steps_per_second": 9.444,
|
| 1959 |
-
"step": 250
|
| 1960 |
-
},
|
| 1961 |
-
{
|
| 1962 |
-
"epoch": 6.972222222222222,
|
| 1963 |
-
"grad_norm": 0.7158817052841187,
|
| 1964 |
-
"learning_rate": 3.114285714285715e-05,
|
| 1965 |
-
"loss": 0.3125,
|
| 1966 |
-
"step": 251
|
| 1967 |
-
},
|
| 1968 |
-
{
|
| 1969 |
-
"epoch": 7.0,
|
| 1970 |
-
"grad_norm": 0.8010092973709106,
|
| 1971 |
-
"learning_rate": 3.0857142857142856e-05,
|
| 1972 |
-
"loss": 0.3094,
|
| 1973 |
-
"step": 252
|
| 1974 |
-
},
|
| 1975 |
-
{
|
| 1976 |
-
"epoch": 7.027777777777778,
|
| 1977 |
-
"grad_norm": 0.7418866157531738,
|
| 1978 |
-
"learning_rate": 3.057142857142857e-05,
|
| 1979 |
-
"loss": 0.3067,
|
| 1980 |
-
"step": 253
|
| 1981 |
-
},
|
| 1982 |
-
{
|
| 1983 |
-
"epoch": 7.055555555555555,
|
| 1984 |
-
"grad_norm": 0.6731083989143372,
|
| 1985 |
-
"learning_rate": 3.0285714285714288e-05,
|
| 1986 |
-
"loss": 0.3068,
|
| 1987 |
-
"step": 254
|
| 1988 |
-
},
|
| 1989 |
-
{
|
| 1990 |
-
"epoch": 7.083333333333333,
|
| 1991 |
-
"grad_norm": 0.6405408382415771,
|
| 1992 |
-
"learning_rate": 3e-05,
|
| 1993 |
-
"loss": 0.3075,
|
| 1994 |
-
"step": 255
|
| 1995 |
-
},
|
| 1996 |
-
{
|
| 1997 |
-
"epoch": 7.111111111111111,
|
| 1998 |
-
"grad_norm": 0.6403458118438721,
|
| 1999 |
-
"learning_rate": 2.9714285714285717e-05,
|
| 2000 |
-
"loss": 0.3096,
|
| 2001 |
-
"step": 256
|
| 2002 |
-
},
|
| 2003 |
-
{
|
| 2004 |
-
"epoch": 7.138888888888889,
|
| 2005 |
-
"grad_norm": 0.7583682537078857,
|
| 2006 |
-
"learning_rate": 2.9428571428571426e-05,
|
| 2007 |
-
"loss": 0.3103,
|
| 2008 |
-
"step": 257
|
| 2009 |
-
},
|
| 2010 |
-
{
|
| 2011 |
-
"epoch": 7.166666666666667,
|
| 2012 |
-
"grad_norm": 0.8137710094451904,
|
| 2013 |
-
"learning_rate": 2.9142857142857146e-05,
|
| 2014 |
-
"loss": 0.3064,
|
| 2015 |
-
"step": 258
|
| 2016 |
-
},
|
| 2017 |
-
{
|
| 2018 |
-
"epoch": 7.194444444444445,
|
| 2019 |
-
"grad_norm": 0.7179896235466003,
|
| 2020 |
-
"learning_rate": 2.885714285714286e-05,
|
| 2021 |
-
"loss": 0.3067,
|
| 2022 |
-
"step": 259
|
| 2023 |
-
},
|
| 2024 |
-
{
|
| 2025 |
-
"epoch": 7.222222222222222,
|
| 2026 |
-
"grad_norm": 0.9344987273216248,
|
| 2027 |
-
"learning_rate": 2.857142857142857e-05,
|
| 2028 |
-
"loss": 0.3081,
|
| 2029 |
-
"step": 260
|
| 2030 |
-
},
|
| 2031 |
-
{
|
| 2032 |
-
"epoch": 7.222222222222222,
|
| 2033 |
-
"eval_loss": 0.33135512471199036,
|
| 2034 |
-
"eval_runtime": 0.1054,
|
| 2035 |
-
"eval_samples_per_second": 436.494,
|
| 2036 |
-
"eval_steps_per_second": 9.489,
|
| 2037 |
-
"step": 260
|
| 2038 |
-
},
|
| 2039 |
-
{
|
| 2040 |
-
"epoch": 7.25,
|
| 2041 |
-
"grad_norm": 0.7846106886863708,
|
| 2042 |
-
"learning_rate": 2.8285714285714287e-05,
|
| 2043 |
-
"loss": 0.3024,
|
| 2044 |
-
"step": 261
|
| 2045 |
-
},
|
| 2046 |
-
{
|
| 2047 |
-
"epoch": 7.277777777777778,
|
| 2048 |
-
"grad_norm": 0.5884716510772705,
|
| 2049 |
-
"learning_rate": 2.8000000000000003e-05,
|
| 2050 |
-
"loss": 0.3062,
|
| 2051 |
-
"step": 262
|
| 2052 |
-
},
|
| 2053 |
-
{
|
| 2054 |
-
"epoch": 7.305555555555555,
|
| 2055 |
-
"grad_norm": 0.7277001738548279,
|
| 2056 |
-
"learning_rate": 2.7714285714285716e-05,
|
| 2057 |
-
"loss": 0.3017,
|
| 2058 |
-
"step": 263
|
| 2059 |
-
},
|
| 2060 |
-
{
|
| 2061 |
-
"epoch": 7.333333333333333,
|
| 2062 |
-
"grad_norm": 0.6671104431152344,
|
| 2063 |
-
"learning_rate": 2.742857142857143e-05,
|
| 2064 |
-
"loss": 0.3109,
|
| 2065 |
-
"step": 264
|
| 2066 |
-
},
|
| 2067 |
-
{
|
| 2068 |
-
"epoch": 7.361111111111111,
|
| 2069 |
-
"grad_norm": 0.6468051671981812,
|
| 2070 |
-
"learning_rate": 2.714285714285714e-05,
|
| 2071 |
-
"loss": 0.3051,
|
| 2072 |
-
"step": 265
|
| 2073 |
-
},
|
| 2074 |
-
{
|
| 2075 |
-
"epoch": 7.388888888888889,
|
| 2076 |
-
"grad_norm": 0.7413132190704346,
|
| 2077 |
-
"learning_rate": 2.6857142857142857e-05,
|
| 2078 |
-
"loss": 0.3059,
|
| 2079 |
-
"step": 266
|
| 2080 |
-
},
|
| 2081 |
-
{
|
| 2082 |
-
"epoch": 7.416666666666667,
|
| 2083 |
-
"grad_norm": 0.8842555284500122,
|
| 2084 |
-
"learning_rate": 2.6571428571428576e-05,
|
| 2085 |
-
"loss": 0.3108,
|
| 2086 |
-
"step": 267
|
| 2087 |
-
},
|
| 2088 |
-
{
|
| 2089 |
-
"epoch": 7.444444444444445,
|
| 2090 |
-
"grad_norm": 0.7701683044433594,
|
| 2091 |
-
"learning_rate": 2.6285714285714286e-05,
|
| 2092 |
-
"loss": 0.31,
|
| 2093 |
-
"step": 268
|
| 2094 |
-
},
|
| 2095 |
-
{
|
| 2096 |
-
"epoch": 7.472222222222222,
|
| 2097 |
-
"grad_norm": 0.6261523962020874,
|
| 2098 |
-
"learning_rate": 2.6000000000000002e-05,
|
| 2099 |
-
"loss": 0.2966,
|
| 2100 |
-
"step": 269
|
| 2101 |
-
},
|
| 2102 |
-
{
|
| 2103 |
-
"epoch": 7.5,
|
| 2104 |
-
"grad_norm": 0.6180337071418762,
|
| 2105 |
-
"learning_rate": 2.5714285714285714e-05,
|
| 2106 |
-
"loss": 0.3063,
|
| 2107 |
-
"step": 270
|
| 2108 |
-
},
|
| 2109 |
-
{
|
| 2110 |
-
"epoch": 7.5,
|
| 2111 |
-
"eval_loss": 0.33175382018089294,
|
| 2112 |
-
"eval_runtime": 0.1062,
|
| 2113 |
-
"eval_samples_per_second": 433.162,
|
| 2114 |
-
"eval_steps_per_second": 9.417,
|
| 2115 |
-
"step": 270
|
| 2116 |
-
},
|
| 2117 |
-
{
|
| 2118 |
-
"epoch": 7.527777777777778,
|
| 2119 |
-
"grad_norm": 0.910743236541748,
|
| 2120 |
-
"learning_rate": 2.542857142857143e-05,
|
| 2121 |
-
"loss": 0.3058,
|
| 2122 |
-
"step": 271
|
| 2123 |
-
},
|
| 2124 |
-
{
|
| 2125 |
-
"epoch": 7.555555555555555,
|
| 2126 |
-
"grad_norm": 0.8947623372077942,
|
| 2127 |
-
"learning_rate": 2.5142857142857147e-05,
|
| 2128 |
-
"loss": 0.3092,
|
| 2129 |
-
"step": 272
|
| 2130 |
-
},
|
| 2131 |
-
{
|
| 2132 |
-
"epoch": 7.583333333333333,
|
| 2133 |
-
"grad_norm": 0.8466401100158691,
|
| 2134 |
-
"learning_rate": 2.485714285714286e-05,
|
| 2135 |
-
"loss": 0.3069,
|
| 2136 |
-
"step": 273
|
| 2137 |
-
},
|
| 2138 |
-
{
|
| 2139 |
-
"epoch": 7.611111111111111,
|
| 2140 |
-
"grad_norm": 0.7808226943016052,
|
| 2141 |
-
"learning_rate": 2.4571428571428572e-05,
|
| 2142 |
-
"loss": 0.316,
|
| 2143 |
-
"step": 274
|
| 2144 |
-
},
|
| 2145 |
-
{
|
| 2146 |
-
"epoch": 7.638888888888889,
|
| 2147 |
-
"grad_norm": 0.6704230308532715,
|
| 2148 |
-
"learning_rate": 2.4285714285714288e-05,
|
| 2149 |
-
"loss": 0.2993,
|
| 2150 |
-
"step": 275
|
| 2151 |
-
},
|
| 2152 |
-
{
|
| 2153 |
-
"epoch": 7.666666666666667,
|
| 2154 |
-
"grad_norm": 0.7090719938278198,
|
| 2155 |
-
"learning_rate": 2.4e-05,
|
| 2156 |
-
"loss": 0.3034,
|
| 2157 |
-
"step": 276
|
| 2158 |
-
},
|
| 2159 |
-
{
|
| 2160 |
-
"epoch": 7.694444444444445,
|
| 2161 |
-
"grad_norm": 0.7552341818809509,
|
| 2162 |
-
"learning_rate": 2.3714285714285717e-05,
|
| 2163 |
-
"loss": 0.3077,
|
| 2164 |
-
"step": 277
|
| 2165 |
-
},
|
| 2166 |
-
{
|
| 2167 |
-
"epoch": 7.722222222222222,
|
| 2168 |
-
"grad_norm": 0.7747870683670044,
|
| 2169 |
-
"learning_rate": 2.342857142857143e-05,
|
| 2170 |
-
"loss": 0.3047,
|
| 2171 |
-
"step": 278
|
| 2172 |
-
},
|
| 2173 |
-
{
|
| 2174 |
-
"epoch": 7.75,
|
| 2175 |
-
"grad_norm": 1.1127567291259766,
|
| 2176 |
-
"learning_rate": 2.3142857142857145e-05,
|
| 2177 |
-
"loss": 0.3105,
|
| 2178 |
-
"step": 279
|
| 2179 |
-
},
|
| 2180 |
-
{
|
| 2181 |
-
"epoch": 7.777777777777778,
|
| 2182 |
-
"grad_norm": 0.7083399891853333,
|
| 2183 |
-
"learning_rate": 2.2857142857142858e-05,
|
| 2184 |
-
"loss": 0.2997,
|
| 2185 |
-
"step": 280
|
| 2186 |
-
},
|
| 2187 |
-
{
|
| 2188 |
-
"epoch": 7.777777777777778,
|
| 2189 |
-
"eval_loss": 0.3296756446361542,
|
| 2190 |
-
"eval_runtime": 0.1057,
|
| 2191 |
-
"eval_samples_per_second": 435.219,
|
| 2192 |
-
"eval_steps_per_second": 9.461,
|
| 2193 |
-
"step": 280
|
| 2194 |
-
},
|
| 2195 |
-
{
|
| 2196 |
-
"epoch": 7.805555555555555,
|
| 2197 |
-
"grad_norm": 0.5560160279273987,
|
| 2198 |
-
"learning_rate": 2.257142857142857e-05,
|
| 2199 |
-
"loss": 0.3035,
|
| 2200 |
-
"step": 281
|
| 2201 |
-
},
|
| 2202 |
-
{
|
| 2203 |
-
"epoch": 7.833333333333333,
|
| 2204 |
-
"grad_norm": 0.7277750372886658,
|
| 2205 |
-
"learning_rate": 2.2285714285714287e-05,
|
| 2206 |
-
"loss": 0.3066,
|
| 2207 |
-
"step": 282
|
| 2208 |
-
},
|
| 2209 |
-
{
|
| 2210 |
-
"epoch": 7.861111111111111,
|
| 2211 |
-
"grad_norm": 0.683189868927002,
|
| 2212 |
-
"learning_rate": 2.2000000000000003e-05,
|
| 2213 |
-
"loss": 0.3049,
|
| 2214 |
-
"step": 283
|
| 2215 |
-
},
|
| 2216 |
-
{
|
| 2217 |
-
"epoch": 7.888888888888889,
|
| 2218 |
-
"grad_norm": 0.7173867225646973,
|
| 2219 |
-
"learning_rate": 2.1714285714285715e-05,
|
| 2220 |
-
"loss": 0.3098,
|
| 2221 |
-
"step": 284
|
| 2222 |
-
},
|
| 2223 |
-
{
|
| 2224 |
-
"epoch": 7.916666666666667,
|
| 2225 |
-
"grad_norm": 0.698379397392273,
|
| 2226 |
-
"learning_rate": 2.1428571428571428e-05,
|
| 2227 |
-
"loss": 0.3081,
|
| 2228 |
-
"step": 285
|
| 2229 |
-
},
|
| 2230 |
-
{
|
| 2231 |
-
"epoch": 7.944444444444445,
|
| 2232 |
-
"grad_norm": 0.5789396166801453,
|
| 2233 |
-
"learning_rate": 2.1142857142857144e-05,
|
| 2234 |
-
"loss": 0.3109,
|
| 2235 |
-
"step": 286
|
| 2236 |
-
},
|
| 2237 |
-
{
|
| 2238 |
-
"epoch": 7.972222222222222,
|
| 2239 |
-
"grad_norm": 0.8007158637046814,
|
| 2240 |
-
"learning_rate": 2.0857142857142857e-05,
|
| 2241 |
-
"loss": 0.3149,
|
| 2242 |
-
"step": 287
|
| 2243 |
-
},
|
| 2244 |
-
{
|
| 2245 |
-
"epoch": 8.0,
|
| 2246 |
-
"grad_norm": 0.8773916959762573,
|
| 2247 |
-
"learning_rate": 2.0571428571428573e-05,
|
| 2248 |
-
"loss": 0.3087,
|
| 2249 |
-
"step": 288
|
| 2250 |
-
},
|
| 2251 |
-
{
|
| 2252 |
-
"epoch": 8.027777777777779,
|
| 2253 |
-
"grad_norm": 0.7212807536125183,
|
| 2254 |
-
"learning_rate": 2.0285714285714286e-05,
|
| 2255 |
-
"loss": 0.2989,
|
| 2256 |
-
"step": 289
|
| 2257 |
-
},
|
| 2258 |
-
{
|
| 2259 |
-
"epoch": 8.055555555555555,
|
| 2260 |
-
"grad_norm": 0.6096076369285583,
|
| 2261 |
-
"learning_rate": 2e-05,
|
| 2262 |
-
"loss": 0.3069,
|
| 2263 |
-
"step": 290
|
| 2264 |
-
},
|
| 2265 |
-
{
|
| 2266 |
-
"epoch": 8.055555555555555,
|
| 2267 |
-
"eval_loss": 0.33069342374801636,
|
| 2268 |
-
"eval_runtime": 0.105,
|
| 2269 |
-
"eval_samples_per_second": 438.062,
|
| 2270 |
-
"eval_steps_per_second": 9.523,
|
| 2271 |
-
"step": 290
|
| 2272 |
-
},
|
| 2273 |
-
{
|
| 2274 |
-
"epoch": 8.083333333333334,
|
| 2275 |
-
"grad_norm": 0.5173125267028809,
|
| 2276 |
-
"learning_rate": 1.9714285714285714e-05,
|
| 2277 |
-
"loss": 0.295,
|
| 2278 |
-
"step": 291
|
| 2279 |
-
},
|
| 2280 |
-
{
|
| 2281 |
-
"epoch": 8.11111111111111,
|
| 2282 |
-
"grad_norm": 0.6913369297981262,
|
| 2283 |
-
"learning_rate": 1.942857142857143e-05,
|
| 2284 |
-
"loss": 0.3014,
|
| 2285 |
-
"step": 292
|
| 2286 |
-
},
|
| 2287 |
-
{
|
| 2288 |
-
"epoch": 8.13888888888889,
|
| 2289 |
-
"grad_norm": 0.7195921540260315,
|
| 2290 |
-
"learning_rate": 1.9142857142857143e-05,
|
| 2291 |
-
"loss": 0.3037,
|
| 2292 |
-
"step": 293
|
| 2293 |
-
},
|
| 2294 |
-
{
|
| 2295 |
-
"epoch": 8.166666666666666,
|
| 2296 |
-
"grad_norm": 0.6366473436355591,
|
| 2297 |
-
"learning_rate": 1.885714285714286e-05,
|
| 2298 |
-
"loss": 0.3031,
|
| 2299 |
-
"step": 294
|
| 2300 |
-
},
|
| 2301 |
-
{
|
| 2302 |
-
"epoch": 8.194444444444445,
|
| 2303 |
-
"grad_norm": 0.5457173585891724,
|
| 2304 |
-
"learning_rate": 1.8571428571428572e-05,
|
| 2305 |
-
"loss": 0.2957,
|
| 2306 |
-
"step": 295
|
| 2307 |
-
},
|
| 2308 |
-
{
|
| 2309 |
-
"epoch": 8.222222222222221,
|
| 2310 |
-
"grad_norm": 0.6149912476539612,
|
| 2311 |
-
"learning_rate": 1.8285714285714288e-05,
|
| 2312 |
-
"loss": 0.2997,
|
| 2313 |
-
"step": 296
|
| 2314 |
-
},
|
| 2315 |
-
{
|
| 2316 |
-
"epoch": 8.25,
|
| 2317 |
-
"grad_norm": 0.5352884531021118,
|
| 2318 |
-
"learning_rate": 1.8e-05,
|
| 2319 |
-
"loss": 0.3048,
|
| 2320 |
-
"step": 297
|
| 2321 |
-
},
|
| 2322 |
-
{
|
| 2323 |
-
"epoch": 8.277777777777779,
|
| 2324 |
-
"grad_norm": 0.6278409361839294,
|
| 2325 |
-
"learning_rate": 1.7714285714285713e-05,
|
| 2326 |
-
"loss": 0.308,
|
| 2327 |
-
"step": 298
|
| 2328 |
-
},
|
| 2329 |
-
{
|
| 2330 |
-
"epoch": 8.305555555555555,
|
| 2331 |
-
"grad_norm": 0.5881698727607727,
|
| 2332 |
-
"learning_rate": 1.742857142857143e-05,
|
| 2333 |
-
"loss": 0.3005,
|
| 2334 |
-
"step": 299
|
| 2335 |
-
},
|
| 2336 |
-
{
|
| 2337 |
-
"epoch": 8.333333333333334,
|
| 2338 |
-
"grad_norm": 0.6125136613845825,
|
| 2339 |
-
"learning_rate": 1.7142857142857145e-05,
|
| 2340 |
-
"loss": 0.302,
|
| 2341 |
-
"step": 300
|
| 2342 |
-
},
|
| 2343 |
-
{
|
| 2344 |
-
"epoch": 8.333333333333334,
|
| 2345 |
-
"eval_loss": 0.3336547017097473,
|
| 2346 |
-
"eval_runtime": 0.1055,
|
| 2347 |
-
"eval_samples_per_second": 436.2,
|
| 2348 |
-
"eval_steps_per_second": 9.483,
|
| 2349 |
-
"step": 300
|
| 2350 |
-
},
|
| 2351 |
-
{
|
| 2352 |
-
"epoch": 8.36111111111111,
|
| 2353 |
-
"grad_norm": 0.6722866892814636,
|
| 2354 |
-
"learning_rate": 1.6857142857142858e-05,
|
| 2355 |
-
"loss": 0.3012,
|
| 2356 |
-
"step": 301
|
| 2357 |
-
},
|
| 2358 |
-
{
|
| 2359 |
-
"epoch": 8.38888888888889,
|
| 2360 |
-
"grad_norm": 0.6827422976493835,
|
| 2361 |
-
"learning_rate": 1.657142857142857e-05,
|
| 2362 |
-
"loss": 0.297,
|
| 2363 |
-
"step": 302
|
| 2364 |
-
},
|
| 2365 |
-
{
|
| 2366 |
-
"epoch": 8.416666666666666,
|
| 2367 |
-
"grad_norm": 0.7612675428390503,
|
| 2368 |
-
"learning_rate": 1.6285714285714287e-05,
|
| 2369 |
-
"loss": 0.2977,
|
| 2370 |
-
"step": 303
|
| 2371 |
-
},
|
| 2372 |
-
{
|
| 2373 |
-
"epoch": 8.444444444444445,
|
| 2374 |
-
"grad_norm": 0.5952971577644348,
|
| 2375 |
-
"learning_rate": 1.6000000000000003e-05,
|
| 2376 |
-
"loss": 0.3009,
|
| 2377 |
-
"step": 304
|
| 2378 |
-
},
|
| 2379 |
-
{
|
| 2380 |
-
"epoch": 8.472222222222221,
|
| 2381 |
-
"grad_norm": 0.8323265314102173,
|
| 2382 |
-
"learning_rate": 1.5714285714285715e-05,
|
| 2383 |
-
"loss": 0.3043,
|
| 2384 |
-
"step": 305
|
| 2385 |
-
},
|
| 2386 |
-
{
|
| 2387 |
-
"epoch": 8.5,
|
| 2388 |
-
"grad_norm": 0.8321357369422913,
|
| 2389 |
-
"learning_rate": 1.5428571428571428e-05,
|
| 2390 |
-
"loss": 0.2956,
|
| 2391 |
-
"step": 306
|
| 2392 |
-
},
|
| 2393 |
-
{
|
| 2394 |
-
"epoch": 8.527777777777779,
|
| 2395 |
-
"grad_norm": 0.6457182168960571,
|
| 2396 |
-
"learning_rate": 1.5142857142857144e-05,
|
| 2397 |
-
"loss": 0.3029,
|
| 2398 |
-
"step": 307
|
| 2399 |
-
},
|
| 2400 |
-
{
|
| 2401 |
-
"epoch": 8.555555555555555,
|
| 2402 |
-
"grad_norm": 0.5753086805343628,
|
| 2403 |
-
"learning_rate": 1.4857142857142858e-05,
|
| 2404 |
-
"loss": 0.2972,
|
| 2405 |
-
"step": 308
|
| 2406 |
-
},
|
| 2407 |
-
{
|
| 2408 |
-
"epoch": 8.583333333333334,
|
| 2409 |
-
"grad_norm": 0.8767444491386414,
|
| 2410 |
-
"learning_rate": 1.4571428571428573e-05,
|
| 2411 |
-
"loss": 0.2966,
|
| 2412 |
-
"step": 309
|
| 2413 |
-
},
|
| 2414 |
-
{
|
| 2415 |
-
"epoch": 8.61111111111111,
|
| 2416 |
-
"grad_norm": 0.929669201374054,
|
| 2417 |
-
"learning_rate": 1.4285714285714285e-05,
|
| 2418 |
-
"loss": 0.3049,
|
| 2419 |
-
"step": 310
|
| 2420 |
-
},
|
| 2421 |
-
{
|
| 2422 |
-
"epoch": 8.61111111111111,
|
| 2423 |
-
"eval_loss": 0.3303754925727844,
|
| 2424 |
-
"eval_runtime": 0.1057,
|
| 2425 |
-
"eval_samples_per_second": 435.106,
|
| 2426 |
-
"eval_steps_per_second": 9.459,
|
| 2427 |
-
"step": 310
|
| 2428 |
-
},
|
| 2429 |
-
{
|
| 2430 |
-
"epoch": 8.63888888888889,
|
| 2431 |
-
"grad_norm": 0.7576697468757629,
|
| 2432 |
-
"learning_rate": 1.4000000000000001e-05,
|
| 2433 |
-
"loss": 0.2989,
|
| 2434 |
-
"step": 311
|
| 2435 |
-
},
|
| 2436 |
-
{
|
| 2437 |
-
"epoch": 8.666666666666666,
|
| 2438 |
-
"grad_norm": 0.6402246952056885,
|
| 2439 |
-
"learning_rate": 1.3714285714285716e-05,
|
| 2440 |
-
"loss": 0.3051,
|
| 2441 |
-
"step": 312
|
| 2442 |
-
},
|
| 2443 |
-
{
|
| 2444 |
-
"epoch": 8.694444444444445,
|
| 2445 |
-
"grad_norm": 0.5665248036384583,
|
| 2446 |
-
"learning_rate": 1.3428571428571429e-05,
|
| 2447 |
-
"loss": 0.2974,
|
| 2448 |
-
"step": 313
|
| 2449 |
-
},
|
| 2450 |
-
{
|
| 2451 |
-
"epoch": 8.722222222222221,
|
| 2452 |
-
"grad_norm": 0.9747456312179565,
|
| 2453 |
-
"learning_rate": 1.3142857142857143e-05,
|
| 2454 |
-
"loss": 0.3061,
|
| 2455 |
-
"step": 314
|
| 2456 |
-
},
|
| 2457 |
-
{
|
| 2458 |
-
"epoch": 8.75,
|
| 2459 |
-
"grad_norm": 0.657123863697052,
|
| 2460 |
-
"learning_rate": 1.2857142857142857e-05,
|
| 2461 |
-
"loss": 0.3012,
|
| 2462 |
-
"step": 315
|
| 2463 |
-
},
|
| 2464 |
-
{
|
| 2465 |
-
"epoch": 8.777777777777779,
|
| 2466 |
-
"grad_norm": 0.7186892032623291,
|
| 2467 |
-
"learning_rate": 1.2571428571428573e-05,
|
| 2468 |
-
"loss": 0.2995,
|
| 2469 |
-
"step": 316
|
| 2470 |
-
},
|
| 2471 |
-
{
|
| 2472 |
-
"epoch": 8.805555555555555,
|
| 2473 |
-
"grad_norm": 0.6889364123344421,
|
| 2474 |
-
"learning_rate": 1.2285714285714286e-05,
|
| 2475 |
-
"loss": 0.3026,
|
| 2476 |
-
"step": 317
|
| 2477 |
-
},
|
| 2478 |
-
{
|
| 2479 |
-
"epoch": 8.833333333333334,
|
| 2480 |
-
"grad_norm": 0.6299145817756653,
|
| 2481 |
-
"learning_rate": 1.2e-05,
|
| 2482 |
-
"loss": 0.3009,
|
| 2483 |
-
"step": 318
|
| 2484 |
-
},
|
| 2485 |
-
{
|
| 2486 |
-
"epoch": 8.86111111111111,
|
| 2487 |
-
"grad_norm": 0.7328559756278992,
|
| 2488 |
-
"learning_rate": 1.1714285714285715e-05,
|
| 2489 |
-
"loss": 0.3002,
|
| 2490 |
-
"step": 319
|
| 2491 |
-
},
|
| 2492 |
-
{
|
| 2493 |
-
"epoch": 8.88888888888889,
|
| 2494 |
-
"grad_norm": 0.6111913919448853,
|
| 2495 |
-
"learning_rate": 1.1428571428571429e-05,
|
| 2496 |
-
"loss": 0.2953,
|
| 2497 |
-
"step": 320
|
| 2498 |
-
},
|
| 2499 |
-
{
|
| 2500 |
-
"epoch": 8.88888888888889,
|
| 2501 |
-
"eval_loss": 0.3253832757472992,
|
| 2502 |
-
"eval_runtime": 0.106,
|
| 2503 |
-
"eval_samples_per_second": 434.082,
|
| 2504 |
-
"eval_steps_per_second": 9.437,
|
| 2505 |
-
"step": 320
|
| 2506 |
-
},
|
| 2507 |
-
{
|
| 2508 |
-
"epoch": 8.916666666666666,
|
| 2509 |
-
"grad_norm": 0.6739629507064819,
|
| 2510 |
-
"learning_rate": 1.1142857142857143e-05,
|
| 2511 |
-
"loss": 0.3023,
|
| 2512 |
-
"step": 321
|
| 2513 |
-
},
|
| 2514 |
-
{
|
| 2515 |
-
"epoch": 8.944444444444445,
|
| 2516 |
-
"grad_norm": 0.6967675685882568,
|
| 2517 |
-
"learning_rate": 1.0857142857142858e-05,
|
| 2518 |
-
"loss": 0.2978,
|
| 2519 |
-
"step": 322
|
| 2520 |
-
},
|
| 2521 |
-
{
|
| 2522 |
-
"epoch": 8.972222222222221,
|
| 2523 |
-
"grad_norm": 0.702989935874939,
|
| 2524 |
-
"learning_rate": 1.0571428571428572e-05,
|
| 2525 |
-
"loss": 0.3043,
|
| 2526 |
-
"step": 323
|
| 2527 |
-
},
|
| 2528 |
-
{
|
| 2529 |
-
"epoch": 9.0,
|
| 2530 |
-
"grad_norm": 1.156525731086731,
|
| 2531 |
-
"learning_rate": 1.0285714285714286e-05,
|
| 2532 |
-
"loss": 0.3034,
|
| 2533 |
-
"step": 324
|
| 2534 |
-
},
|
| 2535 |
-
{
|
| 2536 |
-
"epoch": 9.027777777777779,
|
| 2537 |
-
"grad_norm": 0.564460277557373,
|
| 2538 |
-
"learning_rate": 1e-05,
|
| 2539 |
-
"loss": 0.2989,
|
| 2540 |
-
"step": 325
|
| 2541 |
-
},
|
| 2542 |
-
{
|
| 2543 |
-
"epoch": 9.055555555555555,
|
| 2544 |
-
"grad_norm": 0.5435044169425964,
|
| 2545 |
-
"learning_rate": 9.714285714285715e-06,
|
| 2546 |
-
"loss": 0.2955,
|
| 2547 |
-
"step": 326
|
| 2548 |
-
},
|
| 2549 |
-
{
|
| 2550 |
-
"epoch": 9.083333333333334,
|
| 2551 |
-
"grad_norm": 0.511762797832489,
|
| 2552 |
-
"learning_rate": 9.42857142857143e-06,
|
| 2553 |
-
"loss": 0.2986,
|
| 2554 |
-
"step": 327
|
| 2555 |
-
},
|
| 2556 |
-
{
|
| 2557 |
-
"epoch": 9.11111111111111,
|
| 2558 |
-
"grad_norm": 0.6208844780921936,
|
| 2559 |
-
"learning_rate": 9.142857142857144e-06,
|
| 2560 |
-
"loss": 0.2932,
|
| 2561 |
-
"step": 328
|
| 2562 |
-
},
|
| 2563 |
-
{
|
| 2564 |
-
"epoch": 9.13888888888889,
|
| 2565 |
-
"grad_norm": 0.5209355354309082,
|
| 2566 |
-
"learning_rate": 8.857142857142857e-06,
|
| 2567 |
-
"loss": 0.2932,
|
| 2568 |
-
"step": 329
|
| 2569 |
-
},
|
| 2570 |
-
{
|
| 2571 |
-
"epoch": 9.166666666666666,
|
| 2572 |
-
"grad_norm": 0.5852081775665283,
|
| 2573 |
-
"learning_rate": 8.571428571428573e-06,
|
| 2574 |
-
"loss": 0.302,
|
| 2575 |
-
"step": 330
|
| 2576 |
-
},
|
| 2577 |
-
{
|
| 2578 |
-
"epoch": 9.166666666666666,
|
| 2579 |
-
"eval_loss": 0.32703322172164917,
|
| 2580 |
-
"eval_runtime": 0.1055,
|
| 2581 |
-
"eval_samples_per_second": 435.944,
|
| 2582 |
-
"eval_steps_per_second": 9.477,
|
| 2583 |
-
"step": 330
|
| 2584 |
-
},
|
| 2585 |
-
{
|
| 2586 |
-
"epoch": 9.194444444444445,
|
| 2587 |
-
"grad_norm": 0.6613155603408813,
|
| 2588 |
-
"learning_rate": 8.285714285714285e-06,
|
| 2589 |
-
"loss": 0.2973,
|
| 2590 |
-
"step": 331
|
| 2591 |
-
},
|
| 2592 |
-
{
|
| 2593 |
-
"epoch": 9.222222222222221,
|
| 2594 |
-
"grad_norm": 0.6458805203437805,
|
| 2595 |
-
"learning_rate": 8.000000000000001e-06,
|
| 2596 |
-
"loss": 0.296,
|
| 2597 |
-
"step": 332
|
| 2598 |
-
},
|
| 2599 |
-
{
|
| 2600 |
-
"epoch": 9.25,
|
| 2601 |
-
"grad_norm": 0.5602886080741882,
|
| 2602 |
-
"learning_rate": 7.714285714285714e-06,
|
| 2603 |
-
"loss": 0.2998,
|
| 2604 |
-
"step": 333
|
| 2605 |
-
},
|
| 2606 |
-
{
|
| 2607 |
-
"epoch": 9.277777777777779,
|
| 2608 |
-
"grad_norm": 0.5723817348480225,
|
| 2609 |
-
"learning_rate": 7.428571428571429e-06,
|
| 2610 |
-
"loss": 0.2898,
|
| 2611 |
-
"step": 334
|
| 2612 |
-
},
|
| 2613 |
-
{
|
| 2614 |
-
"epoch": 9.305555555555555,
|
| 2615 |
-
"grad_norm": 0.6257355213165283,
|
| 2616 |
-
"learning_rate": 7.142857142857143e-06,
|
| 2617 |
-
"loss": 0.2935,
|
| 2618 |
-
"step": 335
|
| 2619 |
-
},
|
| 2620 |
-
{
|
| 2621 |
-
"epoch": 9.333333333333334,
|
| 2622 |
-
"grad_norm": 0.6624913811683655,
|
| 2623 |
-
"learning_rate": 6.857142857142858e-06,
|
| 2624 |
-
"loss": 0.2909,
|
| 2625 |
-
"step": 336
|
| 2626 |
-
},
|
| 2627 |
-
{
|
| 2628 |
-
"epoch": 9.36111111111111,
|
| 2629 |
-
"grad_norm": 0.5716632604598999,
|
| 2630 |
-
"learning_rate": 6.5714285714285714e-06,
|
| 2631 |
-
"loss": 0.3001,
|
| 2632 |
-
"step": 337
|
| 2633 |
-
},
|
| 2634 |
-
{
|
| 2635 |
-
"epoch": 9.38888888888889,
|
| 2636 |
-
"grad_norm": 0.6996496319770813,
|
| 2637 |
-
"learning_rate": 6.285714285714287e-06,
|
| 2638 |
-
"loss": 0.2964,
|
| 2639 |
-
"step": 338
|
| 2640 |
-
},
|
| 2641 |
-
{
|
| 2642 |
-
"epoch": 9.416666666666666,
|
| 2643 |
-
"grad_norm": 0.7235862612724304,
|
| 2644 |
-
"learning_rate": 6e-06,
|
| 2645 |
-
"loss": 0.2927,
|
| 2646 |
-
"step": 339
|
| 2647 |
-
},
|
| 2648 |
-
{
|
| 2649 |
-
"epoch": 9.444444444444445,
|
| 2650 |
-
"grad_norm": 0.6455687284469604,
|
| 2651 |
-
"learning_rate": 5.7142857142857145e-06,
|
| 2652 |
-
"loss": 0.2956,
|
| 2653 |
-
"step": 340
|
| 2654 |
-
},
|
| 2655 |
-
{
|
| 2656 |
-
"epoch": 9.444444444444445,
|
| 2657 |
-
"eval_loss": 0.32588478922843933,
|
| 2658 |
-
"eval_runtime": 0.1057,
|
| 2659 |
-
"eval_samples_per_second": 435.067,
|
| 2660 |
-
"eval_steps_per_second": 9.458,
|
| 2661 |
-
"step": 340
|
| 2662 |
-
},
|
| 2663 |
-
{
|
| 2664 |
-
"epoch": 9.472222222222221,
|
| 2665 |
-
"grad_norm": 0.5984405279159546,
|
| 2666 |
-
"learning_rate": 5.428571428571429e-06,
|
| 2667 |
-
"loss": 0.2959,
|
| 2668 |
-
"step": 341
|
| 2669 |
-
},
|
| 2670 |
-
{
|
| 2671 |
-
"epoch": 9.5,
|
| 2672 |
-
"grad_norm": 0.6240008473396301,
|
| 2673 |
-
"learning_rate": 5.142857142857143e-06,
|
| 2674 |
-
"loss": 0.2936,
|
| 2675 |
-
"step": 342
|
| 2676 |
-
},
|
| 2677 |
-
{
|
| 2678 |
-
"epoch": 9.527777777777779,
|
| 2679 |
-
"grad_norm": 0.6871291399002075,
|
| 2680 |
-
"learning_rate": 4.857142857142858e-06,
|
| 2681 |
-
"loss": 0.2987,
|
| 2682 |
-
"step": 343
|
| 2683 |
-
},
|
| 2684 |
-
{
|
| 2685 |
-
"epoch": 9.555555555555555,
|
| 2686 |
-
"grad_norm": 0.6369628310203552,
|
| 2687 |
-
"learning_rate": 4.571428571428572e-06,
|
| 2688 |
-
"loss": 0.2941,
|
| 2689 |
-
"step": 344
|
| 2690 |
-
},
|
| 2691 |
-
{
|
| 2692 |
-
"epoch": 9.583333333333334,
|
| 2693 |
-
"grad_norm": 0.6651211977005005,
|
| 2694 |
-
"learning_rate": 4.285714285714286e-06,
|
| 2695 |
-
"loss": 0.2959,
|
| 2696 |
-
"step": 345
|
| 2697 |
-
},
|
| 2698 |
-
{
|
| 2699 |
-
"epoch": 9.61111111111111,
|
| 2700 |
-
"grad_norm": 0.7005758285522461,
|
| 2701 |
-
"learning_rate": 4.000000000000001e-06,
|
| 2702 |
-
"loss": 0.2991,
|
| 2703 |
-
"step": 346
|
| 2704 |
-
},
|
| 2705 |
-
{
|
| 2706 |
-
"epoch": 9.63888888888889,
|
| 2707 |
-
"grad_norm": 0.5685088634490967,
|
| 2708 |
-
"learning_rate": 3.7142857142857146e-06,
|
| 2709 |
-
"loss": 0.2936,
|
| 2710 |
-
"step": 347
|
| 2711 |
-
},
|
| 2712 |
-
{
|
| 2713 |
-
"epoch": 9.666666666666666,
|
| 2714 |
-
"grad_norm": 0.6322896480560303,
|
| 2715 |
-
"learning_rate": 3.428571428571429e-06,
|
| 2716 |
-
"loss": 0.2947,
|
| 2717 |
-
"step": 348
|
| 2718 |
-
},
|
| 2719 |
-
{
|
| 2720 |
-
"epoch": 9.694444444444445,
|
| 2721 |
-
"grad_norm": 0.6149244904518127,
|
| 2722 |
-
"learning_rate": 3.1428571428571433e-06,
|
| 2723 |
-
"loss": 0.2951,
|
| 2724 |
-
"step": 349
|
| 2725 |
-
},
|
| 2726 |
-
{
|
| 2727 |
-
"epoch": 9.722222222222221,
|
| 2728 |
-
"grad_norm": 0.685043215751648,
|
| 2729 |
-
"learning_rate": 2.8571428571428573e-06,
|
| 2730 |
-
"loss": 0.2975,
|
| 2731 |
-
"step": 350
|
| 2732 |
-
},
|
| 2733 |
-
{
|
| 2734 |
-
"epoch": 9.722222222222221,
|
| 2735 |
-
"eval_loss": 0.3277616798877716,
|
| 2736 |
-
"eval_runtime": 0.1059,
|
| 2737 |
-
"eval_samples_per_second": 434.21,
|
| 2738 |
-
"eval_steps_per_second": 9.439,
|
| 2739 |
-
"step": 350
|
| 2740 |
-
},
|
| 2741 |
-
{
|
| 2742 |
-
"epoch": 9.75,
|
| 2743 |
-
"grad_norm": 0.8039355874061584,
|
| 2744 |
-
"learning_rate": 2.5714285714285716e-06,
|
| 2745 |
-
"loss": 0.2951,
|
| 2746 |
-
"step": 351
|
| 2747 |
-
},
|
| 2748 |
-
{
|
| 2749 |
-
"epoch": 9.777777777777779,
|
| 2750 |
-
"grad_norm": 0.6388571262359619,
|
| 2751 |
-
"learning_rate": 2.285714285714286e-06,
|
| 2752 |
-
"loss": 0.2928,
|
| 2753 |
-
"step": 352
|
| 2754 |
-
},
|
| 2755 |
-
{
|
| 2756 |
-
"epoch": 9.805555555555555,
|
| 2757 |
-
"grad_norm": 0.5934285521507263,
|
| 2758 |
-
"learning_rate": 2.0000000000000003e-06,
|
| 2759 |
-
"loss": 0.2934,
|
| 2760 |
-
"step": 353
|
| 2761 |
-
},
|
| 2762 |
-
{
|
| 2763 |
-
"epoch": 9.833333333333334,
|
| 2764 |
-
"grad_norm": 0.5320731401443481,
|
| 2765 |
-
"learning_rate": 1.7142857142857145e-06,
|
| 2766 |
-
"loss": 0.2952,
|
| 2767 |
-
"step": 354
|
| 2768 |
-
},
|
| 2769 |
-
{
|
| 2770 |
-
"epoch": 9.86111111111111,
|
| 2771 |
-
"grad_norm": 0.6137614846229553,
|
| 2772 |
-
"learning_rate": 1.4285714285714286e-06,
|
| 2773 |
-
"loss": 0.2932,
|
| 2774 |
-
"step": 355
|
| 2775 |
-
},
|
| 2776 |
-
{
|
| 2777 |
-
"epoch": 9.88888888888889,
|
| 2778 |
-
"grad_norm": 0.8172494769096375,
|
| 2779 |
-
"learning_rate": 1.142857142857143e-06,
|
| 2780 |
-
"loss": 0.2944,
|
| 2781 |
-
"step": 356
|
| 2782 |
-
},
|
| 2783 |
-
{
|
| 2784 |
-
"epoch": 9.916666666666666,
|
| 2785 |
-
"grad_norm": 0.6931514739990234,
|
| 2786 |
-
"learning_rate": 8.571428571428572e-07,
|
| 2787 |
-
"loss": 0.2917,
|
| 2788 |
-
"step": 357
|
| 2789 |
-
},
|
| 2790 |
-
{
|
| 2791 |
-
"epoch": 9.944444444444445,
|
| 2792 |
-
"grad_norm": 0.8408763408660889,
|
| 2793 |
-
"learning_rate": 5.714285714285715e-07,
|
| 2794 |
-
"loss": 0.2952,
|
| 2795 |
-
"step": 358
|
| 2796 |
-
},
|
| 2797 |
-
{
|
| 2798 |
-
"epoch": 9.972222222222221,
|
| 2799 |
-
"grad_norm": 0.6687312126159668,
|
| 2800 |
-
"learning_rate": 2.8571428571428575e-07,
|
| 2801 |
-
"loss": 0.2948,
|
| 2802 |
-
"step": 359
|
| 2803 |
-
},
|
| 2804 |
-
{
|
| 2805 |
-
"epoch": 10.0,
|
| 2806 |
-
"grad_norm": 0.8545575737953186,
|
| 2807 |
-
"learning_rate": 0.0,
|
| 2808 |
-
"loss": 0.2944,
|
| 2809 |
-
"step": 360
|
| 2810 |
-
},
|
| 2811 |
-
{
|
| 2812 |
-
"epoch": 10.0,
|
| 2813 |
-
"eval_loss": 0.3285914361476898,
|
| 2814 |
-
"eval_runtime": 0.1058,
|
| 2815 |
-
"eval_samples_per_second": 434.729,
|
| 2816 |
-
"eval_steps_per_second": 9.451,
|
| 2817 |
-
"step": 360
|
| 2818 |
-
}
|
| 2819 |
-
],
|
| 2820 |
-
"logging_steps": 1,
|
| 2821 |
-
"max_steps": 360,
|
| 2822 |
-
"num_input_tokens_seen": 0,
|
| 2823 |
-
"num_train_epochs": 10,
|
| 2824 |
-
"save_steps": 5000,
|
| 2825 |
-
"stateful_callbacks": {
|
| 2826 |
-
"TrainerControl": {
|
| 2827 |
-
"args": {
|
| 2828 |
-
"should_epoch_stop": false,
|
| 2829 |
-
"should_evaluate": false,
|
| 2830 |
-
"should_log": false,
|
| 2831 |
-
"should_save": true,
|
| 2832 |
-
"should_training_stop": true
|
| 2833 |
-
},
|
| 2834 |
-
"attributes": {}
|
| 2835 |
-
}
|
| 2836 |
-
},
|
| 2837 |
-
"total_flos": 3831354556001280.0,
|
| 2838 |
-
"train_batch_size": 128,
|
| 2839 |
-
"trial_name": null,
|
| 2840 |
-
"trial_params": null
|
| 2841 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:e928b416737e3e579d0a564d29d42c7345239d5cd5670a62b4d71260a823a6d5
|
| 3 |
-
size 6008
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/checkpoint-360/vocab.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/config.json
DELETED
|
@@ -1,42 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_name_or_path": "HuggingFaceTB/SmolLM2-135M-Instruct",
|
| 3 |
-
"action_dim": 26,
|
| 4 |
-
"architectures": [
|
| 5 |
-
"LowdimLlamaForCausalLM"
|
| 6 |
-
],
|
| 7 |
-
"attention_bias": false,
|
| 8 |
-
"attention_dropout": 0.0,
|
| 9 |
-
"bos_token_id": 1,
|
| 10 |
-
"eos_token_id": 2,
|
| 11 |
-
"head_dim": 64,
|
| 12 |
-
"hidden_act": "silu",
|
| 13 |
-
"hidden_size": 576,
|
| 14 |
-
"initializer_range": 0.041666666666666664,
|
| 15 |
-
"intermediate_size": 1536,
|
| 16 |
-
"is_llama_config": true,
|
| 17 |
-
"max_position_embeddings": 8192,
|
| 18 |
-
"mlp_bias": false,
|
| 19 |
-
"model_type": "llama_lowdim",
|
| 20 |
-
"num_attention_heads": 9,
|
| 21 |
-
"num_hidden_layers": 30,
|
| 22 |
-
"num_key_value_heads": 3,
|
| 23 |
-
"obs_dim": 46,
|
| 24 |
-
"pad_token_id": 2,
|
| 25 |
-
"pretraining_tp": 1,
|
| 26 |
-
"rms_norm_eps": 1e-05,
|
| 27 |
-
"rope_interleaved": false,
|
| 28 |
-
"rope_scaling": null,
|
| 29 |
-
"rope_theta": 100000,
|
| 30 |
-
"tie_word_embeddings": true,
|
| 31 |
-
"torch_dtype": "float32",
|
| 32 |
-
"transformers.js_config": {
|
| 33 |
-
"kv_cache_dtype": {
|
| 34 |
-
"fp16": "float16",
|
| 35 |
-
"q4f16": "float16"
|
| 36 |
-
}
|
| 37 |
-
},
|
| 38 |
-
"transformers_version": "4.47.1",
|
| 39 |
-
"use_cache": true,
|
| 40 |
-
"use_joint_mlp_projector": true,
|
| 41 |
-
"vocab_size": 49152
|
| 42 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/generation_config.json
DELETED
|
@@ -1,7 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_from_model_config": true,
|
| 3 |
-
"bos_token_id": 1,
|
| 4 |
-
"eos_token_id": 2,
|
| 5 |
-
"pad_token_id": 2,
|
| 6 |
-
"transformers_version": "4.47.1"
|
| 7 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/model.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:532f9325616dbe87be8846d0b03f80c41a2de5d5a05c7397eb15b3484410ede9
|
| 3 |
-
size 539588496
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/normalizer.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:1b3226fbcd5e7e17c04b2cf20350d5f793927d16481bc217684ce129a4aa170a
|
| 3 |
-
size 5666
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/train.log
DELETED
|
@@ -1,11 +0,0 @@
|
|
| 1 |
-
[2026-03-27 16:20:52,880][numexpr.utils][INFO] - Note: NumExpr detected 24 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
|
| 2 |
-
[2026-03-27 16:20:52,880][numexpr.utils][INFO] - NumExpr defaulting to 8 threads.
|
| 3 |
-
[2026-03-27 16:20:55,182][datasets][INFO] - PyTorch version 2.2.2 available.
|
| 4 |
-
[2026-03-27 16:20:55,182][datasets][INFO] - TensorFlow version 2.15.1 available.
|
| 5 |
-
[2026-03-27 16:20:55,182][datasets][INFO] - JAX version 0.4.30 available.
|
| 6 |
-
[2026-03-27 16:20:59,889][datasets.arrow_dataset][WARNING] - Setting TOKENIZERS_PARALLELISM=false for forked processes.
|
| 7 |
-
[2026-03-27 16:21:01,023][datasets.arrow_dataset][WARNING] - Setting TOKENIZERS_PARALLELISM=false for forked processes.
|
| 8 |
-
[2026-03-27 16:21:01,919][root][INFO] - gcc -pthread -B /home/chyang/miniconda3/envs/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /home/chyang/miniconda3/envs/llm-bc/include -I/home/chyang/miniconda3/envs/llm-bc/include -fPIC -O2 -isystem /home/chyang/miniconda3/envs/llm-bc/include -fPIC -c /tmp/tmp7t0nrkuh/test.c -o /tmp/tmp7t0nrkuh/test.o
|
| 9 |
-
[2026-03-27 16:21:01,969][root][INFO] - gcc -pthread -B /home/chyang/miniconda3/envs/llm-bc/compiler_compat /tmp/tmp7t0nrkuh/test.o -laio -o /tmp/tmp7t0nrkuh/a.out
|
| 10 |
-
[2026-03-27 16:21:02,963][root][INFO] - gcc -pthread -B /home/chyang/miniconda3/envs/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /home/chyang/miniconda3/envs/llm-bc/include -I/home/chyang/miniconda3/envs/llm-bc/include -fPIC -O2 -isystem /home/chyang/miniconda3/envs/llm-bc/include -fPIC -c /tmp/tmps_s5vyh9/test.c -o /tmp/tmps_s5vyh9/test.o
|
| 11 |
-
[2026-03-27 16:21:03,018][root][INFO] - gcc -pthread -B /home/chyang/miniconda3/envs/llm-bc/compiler_compat /tmp/tmps_s5vyh9/test.o -L/usr/local/cuda -L/usr/local/cuda/lib64 -lcufile -o /tmp/tmps_s5vyh9/a.out
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/debug-internal.log
DELETED
|
@@ -1,17 +0,0 @@
|
|
| 1 |
-
{"time":"2026-03-27T16:20:56.144221699+08:00","level":"INFO","msg":"using version","core version":"0.18.6"}
|
| 2 |
-
{"time":"2026-03-27T16:20:56.144248456+08:00","level":"INFO","msg":"created symlink","path":"/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug-core.log"}
|
| 3 |
-
{"time":"2026-03-27T16:20:56.25122394+08:00","level":"INFO","msg":"created new stream","id":"nhmfpc2t"}
|
| 4 |
-
{"time":"2026-03-27T16:20:56.25128922+08:00","level":"INFO","msg":"stream: started","id":"nhmfpc2t"}
|
| 5 |
-
{"time":"2026-03-27T16:20:56.251326647+08:00","level":"INFO","msg":"sender: started","stream_id":"nhmfpc2t"}
|
| 6 |
-
{"time":"2026-03-27T16:20:56.251319685+08:00","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"nhmfpc2t"}}
|
| 7 |
-
{"time":"2026-03-27T16:20:56.251303072+08:00","level":"INFO","msg":"handler: started","stream_id":{"value":"nhmfpc2t"}}
|
| 8 |
-
{"time":"2026-03-27T16:20:57.417356613+08:00","level":"INFO","msg":"Starting system monitor"}
|
| 9 |
-
{"time":"2026-03-27T16:25:43.814415705+08:00","level":"INFO","msg":"Stopping system monitor"}
|
| 10 |
-
{"time":"2026-03-27T16:25:43.815067495+08:00","level":"INFO","msg":"Stopped system monitor"}
|
| 11 |
-
{"time":"2026-03-27T16:25:44.815119947+08:00","level":"INFO","msg":"handler: operation stats","stats":{"operations":[{"desc":"uploading wandb-summary.json","runtime_seconds":0.136938035,"progress":"495B/495B"},{"desc":"saving job artifact","runtime_seconds":0.037243753}],"total_operations":2}}
|
| 12 |
-
{"time":"2026-03-27T16:25:49.240469481+08:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
|
| 13 |
-
{"time":"2026-03-27T16:25:50.901076411+08:00","level":"INFO","msg":"stream: closing","id":"nhmfpc2t"}
|
| 14 |
-
{"time":"2026-03-27T16:25:50.901107424+08:00","level":"INFO","msg":"handler: closed","stream_id":{"value":"nhmfpc2t"}}
|
| 15 |
-
{"time":"2026-03-27T16:25:50.901135119+08:00","level":"INFO","msg":"writer: Close: closed","stream_id":{"value":"nhmfpc2t"}}
|
| 16 |
-
{"time":"2026-03-27T16:25:50.901152717+08:00","level":"INFO","msg":"sender: closed","stream_id":"nhmfpc2t"}
|
| 17 |
-
{"time":"2026-03-27T16:25:50.90122351+08:00","level":"INFO","msg":"stream: closed","id":"nhmfpc2t"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/debug.log
DELETED
|
@@ -1,35 +0,0 @@
|
|
| 1 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Current SDK version is 0.18.6
|
| 2 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Configure stats pid to 2703097
|
| 3 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Loading settings from /home/chyang/.config/wandb/settings
|
| 4 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Loading settings from /tmp2/chyang/workspace/LLM-BC/wandb/settings
|
| 5 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
|
| 6 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': 'online', '_disable_service': None}
|
| 7 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train.py', 'program_abspath': '/tmp2/chyang/workspace/LLM-BC/train.py', 'program': '/tmp2/chyang/workspace/LLM-BC/./train.py'}
|
| 8 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Applying login settings: {}
|
| 9 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:_log_setup():533] Logging user logs to /tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug.log
|
| 10 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:_log_setup():534] Logging internal logs to /tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug-internal.log
|
| 11 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:init():619] calling init triggers
|
| 12 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:init():626] wandb.init called with sweep_config: {}
|
| 13 |
-
config: {'name': 'train_llm_lowdim', '_target_': 'llmbc.workspace.train_llm_workspace.TrainLLMWorkspace', 'obs_dim': 46, 'action_dim': 26, 'horizon': 1, 'n_obs_steps': 1, 'n_action_steps': 1, 'task_name': 'adroit-hand-hammer-v1', 'exp_name': 'train llm', 'model_name': 'HuggingFaceTB/SmolLM2-135M-Instruct', 'use_quantization': False, 'lora_config': {'r': 32, 'lora_alpha': 64, 'lora_dropout': 0.05, 'bias': 'none', 'task_type': 'CAUSAL_LM'}, 'dataset': {'test_data_ratio': 0.01}, 'debug': False, 'training': {'seed': 42, 'per_device_train_batch_size': 128, 'per_device_eval_batch_size': 128, 'gradient_accumulation_steps': 1, 'optim': 'paged_adamw_32bit', 'num_train_epochs': 10, 'eval_strategy': 'steps', 'logging_steps': 1, 'warmup_steps': 10, 'logging_strategy': 'steps', 'learning_rate': 0.0001, 'fp16': False, 'bf16': True, 'tf32': True, 'group_by_length': True, 'report_to': 'wandb', 'save_steps': 5000, 'eval_steps': 10, 'use_joint_mlp_projector': True, 'joint_obs_action_mlp_lr': 5e-05}, 'trainer': {'obs_dim': 46, 'action_dim': 26, 'use_joint_mlp_projector': True, 'max_seq_length': 100, 'dataset_text_field': 'text', 'packing': False}, 'logging': {'project': 'llm_module_finetuning', 'resume': True, 'mode': 'online', 'name': '2026.03.27-16.20.52_train_llm_lowdim_adroit-hand-hammer-v1', 'tags': ['train_llm_lowdim', 'adroit-hand-hammer-v1', 'train llm'], 'id': None, 'group': None}, 'multi_run': {'run_dir': 'data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1', 'wandb_name_base': '2026.03.27-16.20.52_train_llm_lowdim_adroit-hand-hammer-v1'}, 'task': {'name': 'adroit-hand-hammer-v1', 'obs_dim': 46, 'action_dim': 26, 'env_runner': {'_target_': 'llmbc.env_runner.adroit_lowdim_runner.AdroitHandLowdimRunner', 'env_name': 'llf-adroit-adroit-hand-hammer-v1', 'n_train': 10, 'n_test': 50, 'n_envs': 10, 'max_steps': 150, 'n_obs_steps': 1, 'n_action_steps': 1, 'instruction_type': 'b', 'feedback_type': ['hp', 'hn', 'fp'], 'visual': False, 'discount': 0.99}, 'dataset': {'_target_': 'llmbc.dataset.adroit_lowdim_dataset.AdroitHandLowdimDataset', 'data_path': 'datasets/adroit-hand-hammer-v1-general.pt', 'data_path2': 'datasets/adroit-hand-hammer-v1.pt', 'horizon': 1, 'pad_before': 0, 'pad_after': 0, 'obs_eef_target': True, 'use_manual_normalizer': False, 'val_ratio': 0.05, 'dummy_normalizer': False}, 'instructor': {'_target_': 'llmbc.translator.instructor.adroit_instructor.adroit_hand_hammer_v1_instructor.AdroitHandHammerV1Instructor'}}, 'llm': {'name': 'HuggingFaceTB/SmolLM2-135M-Instruct', 'model_name': 'SmolLM2-135M-Instruct', 'config_target': 'llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig', 'causal_lm_target': 'llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM', 'use_quantization': False, 'use_joint_mlp_projector': True, 'llm_mode': 'mlp-finetuned', 'finetune_mode': 'orig', 'checkpoint': 'data/outputs/2026.03.27/14.38.20_train_mlp_projector_adroit-hand-hammer-v1/checkpoints/latest.ckpt', 'max_length': 100, 'lora_config': {'r': 32, 'lora_alpha': 64, 'lora_dropout': 0.05, 'bias': 'none', 'task_type': 'CAUSAL_LM'}, 'prompter': {'_target_': 'llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter', 'use_joint_mlp_projector': True}, 'hydra': {'job': {'override_dirname': 'HuggingFaceTB/SmolLM2-135M-Instruct'}, 'run': {'dir': 'data/outputs/2026.03.27/16.20.52_HuggingFaceTB/SmolLM2-135M-Instruct'}}}}
|
| 14 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:init():669] starting backend
|
| 15 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:init():673] sending inform_init request
|
| 16 |
-
2026-03-27 16:20:56,140 INFO MainThread:2703097 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
|
| 17 |
-
2026-03-27 16:20:56,140 INFO MainThread:2703097 [wandb_init.py:init():686] backend started and connected
|
| 18 |
-
2026-03-27 16:20:56,144 INFO MainThread:2703097 [wandb_init.py:init():781] updated telemetry
|
| 19 |
-
2026-03-27 16:20:56,170 INFO MainThread:2703097 [wandb_init.py:init():814] communicating run to backend with 90.0 second timeout
|
| 20 |
-
2026-03-27 16:20:57,414 INFO MainThread:2703097 [wandb_init.py:init():867] starting run threads in backend
|
| 21 |
-
2026-03-27 16:20:57,521 INFO MainThread:2703097 [wandb_run.py:_console_start():2451] atexit reg
|
| 22 |
-
2026-03-27 16:20:57,522 INFO MainThread:2703097 [wandb_run.py:_redirect():2299] redirect: wrap_raw
|
| 23 |
-
2026-03-27 16:20:57,522 INFO MainThread:2703097 [wandb_run.py:_redirect():2364] Wrapping output streams.
|
| 24 |
-
2026-03-27 16:20:57,522 INFO MainThread:2703097 [wandb_run.py:_redirect():2389] Redirects installed.
|
| 25 |
-
2026-03-27 16:20:57,524 INFO MainThread:2703097 [wandb_init.py:init():911] run started, returning control to user process
|
| 26 |
-
2026-03-27 16:21:04,359 INFO MainThread:2703097 [wandb_run.py:_config_callback():1389] config_cb None None {'obs_dim': 46, 'action_dim': 26, 'use_joint_mlp_projector': True, 'vocab_size': 49152, 'max_position_embeddings': 8192, 'hidden_size': 576, 'intermediate_size': 1536, 'num_hidden_layers': 30, 'num_attention_heads': 9, 'num_key_value_heads': 3, 'hidden_act': 'silu', 'initializer_range': 0.041666666666666664, 'rms_norm_eps': 1e-05, 'pretraining_tp': 1, 'use_cache': False, 'rope_theta': 100000, 'rope_scaling': None, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 64, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'bfloat16', 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': None, 'architectures': ['LlamaForCausalLM'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 2, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'HuggingFaceTB/SmolLM2-135M-Instruct', '_attn_implementation_autoset': True, 'transformers_version': '4.47.1', 'is_llama_config': True, 'model_type': 'llama_lowdim', 'rope_interleaved': False, 'transformers.js_config': {'kv_cache_dtype': {'q4f16': 'float16', 'fp16': 'float16'}}, 'output_dir': '/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 128, 'per_device_eval_batch_size': 128, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 0.0001, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 10, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.0, 'warmup_steps': 10, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': '/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/runs/Mar27_16-21-01_A6000-2', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 5000, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': True, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 10, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': '/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'paged_adamw_32bit', 'optim_args': None, 'adafactor': False, 'group_by_length': True, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'evaluation_strategy': None, 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'dispatch_batches': None, 'split_batches': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'eval_use_gather_object': False, 'average_tokens_across_devices': False, 'dataset_text_field': 'text', 'packing': False, 'max_seq_length': 100, 'dataset_num_proc': None, 'dataset_batch_size': 1000, 'model_init_kwargs': None, 'dataset_kwargs': {}, 'eval_packing': None, 'num_of_sequences': 1024, 'chars_per_token': '<CHARS_PER_TOKEN>', 'use_liger': False, 'joint_obs_action_mlp_lr': 5e-05, 'obs_mlp_lr': None, 'action_mlp_lr': None}
|
| 27 |
-
2026-03-27 16:21:04,361 INFO MainThread:2703097 [wandb_config.py:__setitem__():154] config set model/num_parameters = 134889408 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7e6758745670>>
|
| 28 |
-
2026-03-27 16:21:04,361 INFO MainThread:2703097 [wandb_run.py:_config_callback():1389] config_cb model/num_parameters 134889408 None
|
| 29 |
-
2026-03-27 16:25:43,813 INFO MainThread:2703097 [wandb_run.py:_finish():2146] finishing run chyang25-national-taiwan-university/llm_module_finetuning/nhmfpc2t
|
| 30 |
-
2026-03-27 16:25:43,813 INFO MainThread:2703097 [wandb_run.py:_atexit_cleanup():2414] got exitcode: 0
|
| 31 |
-
2026-03-27 16:25:43,813 INFO MainThread:2703097 [wandb_run.py:_restore():2396] restore
|
| 32 |
-
2026-03-27 16:25:43,814 INFO MainThread:2703097 [wandb_run.py:_restore():2402] restore done
|
| 33 |
-
2026-03-27 16:25:50,896 INFO MainThread:2703097 [wandb_run.py:_footer_history_summary_info():3963] rendering history
|
| 34 |
-
2026-03-27 16:25:50,896 INFO MainThread:2703097 [wandb_run.py:_footer_history_summary_info():3995] rendering summary
|
| 35 |
-
2026-03-27 16:25:50,900 INFO MainThread:2703097 [wandb_run.py:_footer_sync_info():3922] logging synced files
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/files/config.yaml
DELETED
|
@@ -1,711 +0,0 @@
|
|
| 1 |
-
_attn_implementation_autoset:
|
| 2 |
-
value: true
|
| 3 |
-
_name_or_path:
|
| 4 |
-
value: HuggingFaceTB/SmolLM2-135M-Instruct
|
| 5 |
-
_target_:
|
| 6 |
-
value: llmbc.workspace.train_llm_workspace.TrainLLMWorkspace
|
| 7 |
-
_wandb:
|
| 8 |
-
value:
|
| 9 |
-
cli_version: 0.18.6
|
| 10 |
-
m:
|
| 11 |
-
- "1": train/loss
|
| 12 |
-
"5": 2
|
| 13 |
-
"6":
|
| 14 |
-
- 1
|
| 15 |
-
- 3
|
| 16 |
-
"7": []
|
| 17 |
-
- "1": train/global_step
|
| 18 |
-
"6":
|
| 19 |
-
- 3
|
| 20 |
-
"7": []
|
| 21 |
-
- "1": train/learning_rate
|
| 22 |
-
"5": 2
|
| 23 |
-
"6":
|
| 24 |
-
- 1
|
| 25 |
-
- 3
|
| 26 |
-
"7": []
|
| 27 |
-
- "1": train/epoch
|
| 28 |
-
"5": 2
|
| 29 |
-
"6":
|
| 30 |
-
- 1
|
| 31 |
-
- 3
|
| 32 |
-
"7": []
|
| 33 |
-
- "1": train/grad_norm
|
| 34 |
-
"5": 2
|
| 35 |
-
"6":
|
| 36 |
-
- 1
|
| 37 |
-
- 3
|
| 38 |
-
"7": []
|
| 39 |
-
- "1": eval/loss
|
| 40 |
-
"5": 2
|
| 41 |
-
"6":
|
| 42 |
-
- 1
|
| 43 |
-
- 3
|
| 44 |
-
"7": []
|
| 45 |
-
- "1": eval/runtime
|
| 46 |
-
"5": 2
|
| 47 |
-
"6":
|
| 48 |
-
- 1
|
| 49 |
-
- 3
|
| 50 |
-
"7": []
|
| 51 |
-
- "1": eval/samples_per_second
|
| 52 |
-
"5": 2
|
| 53 |
-
"6":
|
| 54 |
-
- 1
|
| 55 |
-
- 3
|
| 56 |
-
"7": []
|
| 57 |
-
- "1": eval/steps_per_second
|
| 58 |
-
"5": 2
|
| 59 |
-
"6":
|
| 60 |
-
- 1
|
| 61 |
-
- 3
|
| 62 |
-
"7": []
|
| 63 |
-
python_version: 3.9.20
|
| 64 |
-
t:
|
| 65 |
-
"1":
|
| 66 |
-
- 1
|
| 67 |
-
- 2
|
| 68 |
-
- 3
|
| 69 |
-
- 5
|
| 70 |
-
- 11
|
| 71 |
-
- 12
|
| 72 |
-
- 41
|
| 73 |
-
- 49
|
| 74 |
-
- 50
|
| 75 |
-
- 51
|
| 76 |
-
- 53
|
| 77 |
-
- 55
|
| 78 |
-
- 71
|
| 79 |
-
- 84
|
| 80 |
-
- 98
|
| 81 |
-
"2":
|
| 82 |
-
- 1
|
| 83 |
-
- 2
|
| 84 |
-
- 3
|
| 85 |
-
- 5
|
| 86 |
-
- 11
|
| 87 |
-
- 12
|
| 88 |
-
- 41
|
| 89 |
-
- 49
|
| 90 |
-
- 50
|
| 91 |
-
- 51
|
| 92 |
-
- 53
|
| 93 |
-
- 55
|
| 94 |
-
- 71
|
| 95 |
-
- 84
|
| 96 |
-
- 98
|
| 97 |
-
"3":
|
| 98 |
-
- 2
|
| 99 |
-
- 7
|
| 100 |
-
- 13
|
| 101 |
-
- 15
|
| 102 |
-
- 16
|
| 103 |
-
- 19
|
| 104 |
-
- 23
|
| 105 |
-
- 55
|
| 106 |
-
- 62
|
| 107 |
-
- 66
|
| 108 |
-
"4": 3.9.20
|
| 109 |
-
"5": 0.18.6
|
| 110 |
-
"6": 4.47.1
|
| 111 |
-
"8":
|
| 112 |
-
- 5
|
| 113 |
-
"9":
|
| 114 |
-
"1": transformers_trainer
|
| 115 |
-
"12": 0.18.6
|
| 116 |
-
"13": linux-x86_64
|
| 117 |
-
accelerator_config:
|
| 118 |
-
value:
|
| 119 |
-
dispatch_batches: null
|
| 120 |
-
even_batches: true
|
| 121 |
-
gradient_accumulation_kwargs: null
|
| 122 |
-
non_blocking: false
|
| 123 |
-
split_batches: false
|
| 124 |
-
use_seedable_sampler: true
|
| 125 |
-
action_dim:
|
| 126 |
-
value: 26
|
| 127 |
-
action_mlp_lr:
|
| 128 |
-
value: null
|
| 129 |
-
adafactor:
|
| 130 |
-
value: false
|
| 131 |
-
adam_beta1:
|
| 132 |
-
value: 0.9
|
| 133 |
-
adam_beta2:
|
| 134 |
-
value: 0.999
|
| 135 |
-
adam_epsilon:
|
| 136 |
-
value: 1e-08
|
| 137 |
-
add_cross_attention:
|
| 138 |
-
value: false
|
| 139 |
-
architectures:
|
| 140 |
-
value:
|
| 141 |
-
- LlamaForCausalLM
|
| 142 |
-
attention_bias:
|
| 143 |
-
value: false
|
| 144 |
-
attention_dropout:
|
| 145 |
-
value: 0
|
| 146 |
-
auto_find_batch_size:
|
| 147 |
-
value: false
|
| 148 |
-
average_tokens_across_devices:
|
| 149 |
-
value: false
|
| 150 |
-
bad_words_ids:
|
| 151 |
-
value: null
|
| 152 |
-
batch_eval_metrics:
|
| 153 |
-
value: false
|
| 154 |
-
begin_suppress_tokens:
|
| 155 |
-
value: null
|
| 156 |
-
bf16:
|
| 157 |
-
value: true
|
| 158 |
-
bf16_full_eval:
|
| 159 |
-
value: false
|
| 160 |
-
bos_token_id:
|
| 161 |
-
value: 1
|
| 162 |
-
chars_per_token:
|
| 163 |
-
value: <CHARS_PER_TOKEN>
|
| 164 |
-
chunk_size_feed_forward:
|
| 165 |
-
value: 0
|
| 166 |
-
cross_attention_hidden_size:
|
| 167 |
-
value: null
|
| 168 |
-
data_seed:
|
| 169 |
-
value: null
|
| 170 |
-
dataloader_drop_last:
|
| 171 |
-
value: false
|
| 172 |
-
dataloader_num_workers:
|
| 173 |
-
value: 0
|
| 174 |
-
dataloader_persistent_workers:
|
| 175 |
-
value: false
|
| 176 |
-
dataloader_pin_memory:
|
| 177 |
-
value: true
|
| 178 |
-
dataloader_prefetch_factor:
|
| 179 |
-
value: null
|
| 180 |
-
dataset:
|
| 181 |
-
value:
|
| 182 |
-
test_data_ratio: 0.01
|
| 183 |
-
dataset_batch_size:
|
| 184 |
-
value: 1000
|
| 185 |
-
dataset_num_proc:
|
| 186 |
-
value: null
|
| 187 |
-
dataset_text_field:
|
| 188 |
-
value: text
|
| 189 |
-
ddp_backend:
|
| 190 |
-
value: null
|
| 191 |
-
ddp_broadcast_buffers:
|
| 192 |
-
value: null
|
| 193 |
-
ddp_bucket_cap_mb:
|
| 194 |
-
value: null
|
| 195 |
-
ddp_find_unused_parameters:
|
| 196 |
-
value: null
|
| 197 |
-
ddp_timeout:
|
| 198 |
-
value: 1800
|
| 199 |
-
debug:
|
| 200 |
-
value: []
|
| 201 |
-
decoder_start_token_id:
|
| 202 |
-
value: null
|
| 203 |
-
deepspeed:
|
| 204 |
-
value: null
|
| 205 |
-
disable_tqdm:
|
| 206 |
-
value: false
|
| 207 |
-
dispatch_batches:
|
| 208 |
-
value: null
|
| 209 |
-
diversity_penalty:
|
| 210 |
-
value: 0
|
| 211 |
-
do_eval:
|
| 212 |
-
value: true
|
| 213 |
-
do_predict:
|
| 214 |
-
value: false
|
| 215 |
-
do_sample:
|
| 216 |
-
value: false
|
| 217 |
-
do_train:
|
| 218 |
-
value: false
|
| 219 |
-
early_stopping:
|
| 220 |
-
value: false
|
| 221 |
-
encoder_no_repeat_ngram_size:
|
| 222 |
-
value: 0
|
| 223 |
-
eos_token_id:
|
| 224 |
-
value: 2
|
| 225 |
-
eval_accumulation_steps:
|
| 226 |
-
value: null
|
| 227 |
-
eval_delay:
|
| 228 |
-
value: 0
|
| 229 |
-
eval_do_concat_batches:
|
| 230 |
-
value: true
|
| 231 |
-
eval_on_start:
|
| 232 |
-
value: false
|
| 233 |
-
eval_packing:
|
| 234 |
-
value: null
|
| 235 |
-
eval_steps:
|
| 236 |
-
value: 10
|
| 237 |
-
eval_strategy:
|
| 238 |
-
value: steps
|
| 239 |
-
eval_use_gather_object:
|
| 240 |
-
value: false
|
| 241 |
-
evaluation_strategy:
|
| 242 |
-
value: null
|
| 243 |
-
exp_name:
|
| 244 |
-
value: train llm
|
| 245 |
-
exponential_decay_length_penalty:
|
| 246 |
-
value: null
|
| 247 |
-
finetuning_task:
|
| 248 |
-
value: null
|
| 249 |
-
forced_bos_token_id:
|
| 250 |
-
value: null
|
| 251 |
-
forced_eos_token_id:
|
| 252 |
-
value: null
|
| 253 |
-
fp16:
|
| 254 |
-
value: false
|
| 255 |
-
fp16_backend:
|
| 256 |
-
value: auto
|
| 257 |
-
fp16_full_eval:
|
| 258 |
-
value: false
|
| 259 |
-
fp16_opt_level:
|
| 260 |
-
value: O1
|
| 261 |
-
fsdp:
|
| 262 |
-
value: []
|
| 263 |
-
fsdp_config:
|
| 264 |
-
value:
|
| 265 |
-
min_num_params: 0
|
| 266 |
-
xla: false
|
| 267 |
-
xla_fsdp_grad_ckpt: false
|
| 268 |
-
xla_fsdp_v2: false
|
| 269 |
-
fsdp_min_num_params:
|
| 270 |
-
value: 0
|
| 271 |
-
fsdp_transformer_layer_cls_to_wrap:
|
| 272 |
-
value: null
|
| 273 |
-
full_determinism:
|
| 274 |
-
value: false
|
| 275 |
-
gradient_accumulation_steps:
|
| 276 |
-
value: 1
|
| 277 |
-
gradient_checkpointing:
|
| 278 |
-
value: false
|
| 279 |
-
gradient_checkpointing_kwargs:
|
| 280 |
-
value: null
|
| 281 |
-
greater_is_better:
|
| 282 |
-
value: null
|
| 283 |
-
group_by_length:
|
| 284 |
-
value: true
|
| 285 |
-
half_precision_backend:
|
| 286 |
-
value: auto
|
| 287 |
-
head_dim:
|
| 288 |
-
value: 64
|
| 289 |
-
hidden_act:
|
| 290 |
-
value: silu
|
| 291 |
-
hidden_size:
|
| 292 |
-
value: 576
|
| 293 |
-
horizon:
|
| 294 |
-
value: 1
|
| 295 |
-
hub_always_push:
|
| 296 |
-
value: false
|
| 297 |
-
hub_model_id:
|
| 298 |
-
value: null
|
| 299 |
-
hub_private_repo:
|
| 300 |
-
value: null
|
| 301 |
-
hub_strategy:
|
| 302 |
-
value: every_save
|
| 303 |
-
hub_token:
|
| 304 |
-
value: <HUB_TOKEN>
|
| 305 |
-
id2label:
|
| 306 |
-
value:
|
| 307 |
-
"0": LABEL_0
|
| 308 |
-
"1": LABEL_1
|
| 309 |
-
ignore_data_skip:
|
| 310 |
-
value: false
|
| 311 |
-
include_for_metrics:
|
| 312 |
-
value: []
|
| 313 |
-
include_inputs_for_metrics:
|
| 314 |
-
value: false
|
| 315 |
-
include_num_input_tokens_seen:
|
| 316 |
-
value: false
|
| 317 |
-
include_tokens_per_second:
|
| 318 |
-
value: false
|
| 319 |
-
initializer_range:
|
| 320 |
-
value: 0.041666666666666664
|
| 321 |
-
intermediate_size:
|
| 322 |
-
value: 1536
|
| 323 |
-
is_decoder:
|
| 324 |
-
value: false
|
| 325 |
-
is_encoder_decoder:
|
| 326 |
-
value: false
|
| 327 |
-
is_llama_config:
|
| 328 |
-
value: true
|
| 329 |
-
jit_mode_eval:
|
| 330 |
-
value: false
|
| 331 |
-
joint_obs_action_mlp_lr:
|
| 332 |
-
value: 5e-05
|
| 333 |
-
label_names:
|
| 334 |
-
value: null
|
| 335 |
-
label_smoothing_factor:
|
| 336 |
-
value: 0
|
| 337 |
-
label2id:
|
| 338 |
-
value:
|
| 339 |
-
LABEL_0: 0
|
| 340 |
-
LABEL_1: 1
|
| 341 |
-
learning_rate:
|
| 342 |
-
value: 0.0001
|
| 343 |
-
length_column_name:
|
| 344 |
-
value: length
|
| 345 |
-
length_penalty:
|
| 346 |
-
value: 1
|
| 347 |
-
llm:
|
| 348 |
-
value:
|
| 349 |
-
causal_lm_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM
|
| 350 |
-
checkpoint: data/outputs/2026.03.27/14.38.20_train_mlp_projector_adroit-hand-hammer-v1/checkpoints/latest.ckpt
|
| 351 |
-
config_target: llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig
|
| 352 |
-
finetune_mode: orig
|
| 353 |
-
hydra:
|
| 354 |
-
job:
|
| 355 |
-
override_dirname: HuggingFaceTB/SmolLM2-135M-Instruct
|
| 356 |
-
run:
|
| 357 |
-
dir: data/outputs/2026.03.27/16.20.52_HuggingFaceTB/SmolLM2-135M-Instruct
|
| 358 |
-
llm_mode: mlp-finetuned
|
| 359 |
-
lora_config:
|
| 360 |
-
bias: none
|
| 361 |
-
lora_alpha: 64
|
| 362 |
-
lora_dropout: 0.05
|
| 363 |
-
r: 32
|
| 364 |
-
task_type: CAUSAL_LM
|
| 365 |
-
max_length: 100
|
| 366 |
-
model_name: SmolLM2-135M-Instruct
|
| 367 |
-
name: HuggingFaceTB/SmolLM2-135M-Instruct
|
| 368 |
-
prompter:
|
| 369 |
-
_target_: llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter
|
| 370 |
-
use_joint_mlp_projector: true
|
| 371 |
-
use_joint_mlp_projector: true
|
| 372 |
-
use_quantization: false
|
| 373 |
-
load_best_model_at_end:
|
| 374 |
-
value: false
|
| 375 |
-
local_rank:
|
| 376 |
-
value: 0
|
| 377 |
-
log_level:
|
| 378 |
-
value: passive
|
| 379 |
-
log_level_replica:
|
| 380 |
-
value: warning
|
| 381 |
-
log_on_each_node:
|
| 382 |
-
value: true
|
| 383 |
-
logging:
|
| 384 |
-
value:
|
| 385 |
-
group: null
|
| 386 |
-
id: null
|
| 387 |
-
mode: online
|
| 388 |
-
name: 2026.03.27-16.20.52_train_llm_lowdim_adroit-hand-hammer-v1
|
| 389 |
-
project: llm_module_finetuning
|
| 390 |
-
resume: true
|
| 391 |
-
tags:
|
| 392 |
-
- train_llm_lowdim
|
| 393 |
-
- adroit-hand-hammer-v1
|
| 394 |
-
- train llm
|
| 395 |
-
logging_dir:
|
| 396 |
-
value: /tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/runs/Mar27_16-21-01_A6000-2
|
| 397 |
-
logging_first_step:
|
| 398 |
-
value: false
|
| 399 |
-
logging_nan_inf_filter:
|
| 400 |
-
value: true
|
| 401 |
-
logging_steps:
|
| 402 |
-
value: 1
|
| 403 |
-
logging_strategy:
|
| 404 |
-
value: steps
|
| 405 |
-
lora_config:
|
| 406 |
-
value:
|
| 407 |
-
bias: none
|
| 408 |
-
lora_alpha: 64
|
| 409 |
-
lora_dropout: 0.05
|
| 410 |
-
r: 32
|
| 411 |
-
task_type: CAUSAL_LM
|
| 412 |
-
lr_scheduler_type:
|
| 413 |
-
value: linear
|
| 414 |
-
max_grad_norm:
|
| 415 |
-
value: 1
|
| 416 |
-
max_length:
|
| 417 |
-
value: 20
|
| 418 |
-
max_position_embeddings:
|
| 419 |
-
value: 8192
|
| 420 |
-
max_seq_length:
|
| 421 |
-
value: 100
|
| 422 |
-
max_steps:
|
| 423 |
-
value: -1
|
| 424 |
-
metric_for_best_model:
|
| 425 |
-
value: null
|
| 426 |
-
min_length:
|
| 427 |
-
value: 0
|
| 428 |
-
mlp_bias:
|
| 429 |
-
value: false
|
| 430 |
-
model/num_parameters:
|
| 431 |
-
value: 134889408
|
| 432 |
-
model_init_kwargs:
|
| 433 |
-
value: null
|
| 434 |
-
model_name:
|
| 435 |
-
value: HuggingFaceTB/SmolLM2-135M-Instruct
|
| 436 |
-
model_type:
|
| 437 |
-
value: llama_lowdim
|
| 438 |
-
mp_parameters:
|
| 439 |
-
value: ""
|
| 440 |
-
multi_run:
|
| 441 |
-
value:
|
| 442 |
-
run_dir: data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1
|
| 443 |
-
wandb_name_base: 2026.03.27-16.20.52_train_llm_lowdim_adroit-hand-hammer-v1
|
| 444 |
-
n_action_steps:
|
| 445 |
-
value: 1
|
| 446 |
-
n_obs_steps:
|
| 447 |
-
value: 1
|
| 448 |
-
name:
|
| 449 |
-
value: train_llm_lowdim
|
| 450 |
-
neftune_noise_alpha:
|
| 451 |
-
value: null
|
| 452 |
-
no_cuda:
|
| 453 |
-
value: false
|
| 454 |
-
no_repeat_ngram_size:
|
| 455 |
-
value: 0
|
| 456 |
-
num_attention_heads:
|
| 457 |
-
value: 9
|
| 458 |
-
num_beam_groups:
|
| 459 |
-
value: 1
|
| 460 |
-
num_beams:
|
| 461 |
-
value: 1
|
| 462 |
-
num_hidden_layers:
|
| 463 |
-
value: 30
|
| 464 |
-
num_key_value_heads:
|
| 465 |
-
value: 3
|
| 466 |
-
num_of_sequences:
|
| 467 |
-
value: 1024
|
| 468 |
-
num_return_sequences:
|
| 469 |
-
value: 1
|
| 470 |
-
num_train_epochs:
|
| 471 |
-
value: 10
|
| 472 |
-
obs_dim:
|
| 473 |
-
value: 46
|
| 474 |
-
obs_mlp_lr:
|
| 475 |
-
value: null
|
| 476 |
-
optim:
|
| 477 |
-
value: paged_adamw_32bit
|
| 478 |
-
optim_args:
|
| 479 |
-
value: null
|
| 480 |
-
optim_target_modules:
|
| 481 |
-
value: null
|
| 482 |
-
output_attentions:
|
| 483 |
-
value: false
|
| 484 |
-
output_dir:
|
| 485 |
-
value: /tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1
|
| 486 |
-
output_hidden_states:
|
| 487 |
-
value: false
|
| 488 |
-
output_scores:
|
| 489 |
-
value: false
|
| 490 |
-
overwrite_output_dir:
|
| 491 |
-
value: false
|
| 492 |
-
packing:
|
| 493 |
-
value: false
|
| 494 |
-
pad_token_id:
|
| 495 |
-
value: 2
|
| 496 |
-
past_index:
|
| 497 |
-
value: -1
|
| 498 |
-
per_device_eval_batch_size:
|
| 499 |
-
value: 128
|
| 500 |
-
per_device_train_batch_size:
|
| 501 |
-
value: 128
|
| 502 |
-
per_gpu_eval_batch_size:
|
| 503 |
-
value: null
|
| 504 |
-
per_gpu_train_batch_size:
|
| 505 |
-
value: null
|
| 506 |
-
prediction_loss_only:
|
| 507 |
-
value: false
|
| 508 |
-
prefix:
|
| 509 |
-
value: null
|
| 510 |
-
pretraining_tp:
|
| 511 |
-
value: 1
|
| 512 |
-
problem_type:
|
| 513 |
-
value: null
|
| 514 |
-
push_to_hub:
|
| 515 |
-
value: false
|
| 516 |
-
push_to_hub_model_id:
|
| 517 |
-
value: null
|
| 518 |
-
push_to_hub_organization:
|
| 519 |
-
value: null
|
| 520 |
-
push_to_hub_token:
|
| 521 |
-
value: <PUSH_TO_HUB_TOKEN>
|
| 522 |
-
ray_scope:
|
| 523 |
-
value: last
|
| 524 |
-
remove_invalid_values:
|
| 525 |
-
value: false
|
| 526 |
-
remove_unused_columns:
|
| 527 |
-
value: true
|
| 528 |
-
repetition_penalty:
|
| 529 |
-
value: 1
|
| 530 |
-
report_to:
|
| 531 |
-
value:
|
| 532 |
-
- wandb
|
| 533 |
-
restore_callback_states_from_checkpoint:
|
| 534 |
-
value: false
|
| 535 |
-
resume_from_checkpoint:
|
| 536 |
-
value: null
|
| 537 |
-
return_dict:
|
| 538 |
-
value: true
|
| 539 |
-
return_dict_in_generate:
|
| 540 |
-
value: false
|
| 541 |
-
rms_norm_eps:
|
| 542 |
-
value: 1e-05
|
| 543 |
-
rope_interleaved:
|
| 544 |
-
value: false
|
| 545 |
-
rope_scaling:
|
| 546 |
-
value: null
|
| 547 |
-
rope_theta:
|
| 548 |
-
value: 100000
|
| 549 |
-
run_name:
|
| 550 |
-
value: /tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1
|
| 551 |
-
save_on_each_node:
|
| 552 |
-
value: false
|
| 553 |
-
save_only_model:
|
| 554 |
-
value: false
|
| 555 |
-
save_safetensors:
|
| 556 |
-
value: true
|
| 557 |
-
save_steps:
|
| 558 |
-
value: 5000
|
| 559 |
-
save_strategy:
|
| 560 |
-
value: steps
|
| 561 |
-
save_total_limit:
|
| 562 |
-
value: null
|
| 563 |
-
seed:
|
| 564 |
-
value: 42
|
| 565 |
-
sep_token_id:
|
| 566 |
-
value: null
|
| 567 |
-
skip_memory_metrics:
|
| 568 |
-
value: true
|
| 569 |
-
split_batches:
|
| 570 |
-
value: null
|
| 571 |
-
suppress_tokens:
|
| 572 |
-
value: null
|
| 573 |
-
task:
|
| 574 |
-
value:
|
| 575 |
-
action_dim: 26
|
| 576 |
-
dataset:
|
| 577 |
-
_target_: llmbc.dataset.adroit_lowdim_dataset.AdroitHandLowdimDataset
|
| 578 |
-
data_path: datasets/adroit-hand-hammer-v1-general.pt
|
| 579 |
-
data_path2: datasets/adroit-hand-hammer-v1.pt
|
| 580 |
-
dummy_normalizer: false
|
| 581 |
-
horizon: 1
|
| 582 |
-
obs_eef_target: true
|
| 583 |
-
pad_after: 0
|
| 584 |
-
pad_before: 0
|
| 585 |
-
use_manual_normalizer: false
|
| 586 |
-
val_ratio: 0.05
|
| 587 |
-
env_runner:
|
| 588 |
-
_target_: llmbc.env_runner.adroit_lowdim_runner.AdroitHandLowdimRunner
|
| 589 |
-
discount: 0.99
|
| 590 |
-
env_name: llf-adroit-adroit-hand-hammer-v1
|
| 591 |
-
feedback_type:
|
| 592 |
-
- hp
|
| 593 |
-
- hn
|
| 594 |
-
- fp
|
| 595 |
-
instruction_type: b
|
| 596 |
-
max_steps: 150
|
| 597 |
-
n_action_steps: 1
|
| 598 |
-
n_envs: 10
|
| 599 |
-
n_obs_steps: 1
|
| 600 |
-
n_test: 50
|
| 601 |
-
n_train: 10
|
| 602 |
-
visual: false
|
| 603 |
-
instructor:
|
| 604 |
-
_target_: llmbc.translator.instructor.adroit_instructor.adroit_hand_hammer_v1_instructor.AdroitHandHammerV1Instructor
|
| 605 |
-
name: adroit-hand-hammer-v1
|
| 606 |
-
obs_dim: 46
|
| 607 |
-
task_name:
|
| 608 |
-
value: adroit-hand-hammer-v1
|
| 609 |
-
task_specific_params:
|
| 610 |
-
value: null
|
| 611 |
-
temperature:
|
| 612 |
-
value: 1
|
| 613 |
-
tf_legacy_loss:
|
| 614 |
-
value: false
|
| 615 |
-
tf32:
|
| 616 |
-
value: true
|
| 617 |
-
tie_encoder_decoder:
|
| 618 |
-
value: false
|
| 619 |
-
tie_word_embeddings:
|
| 620 |
-
value: true
|
| 621 |
-
tokenizer_class:
|
| 622 |
-
value: null
|
| 623 |
-
top_k:
|
| 624 |
-
value: 50
|
| 625 |
-
top_p:
|
| 626 |
-
value: 1
|
| 627 |
-
torch_compile:
|
| 628 |
-
value: false
|
| 629 |
-
torch_compile_backend:
|
| 630 |
-
value: null
|
| 631 |
-
torch_compile_mode:
|
| 632 |
-
value: null
|
| 633 |
-
torch_dtype:
|
| 634 |
-
value: bfloat16
|
| 635 |
-
torch_empty_cache_steps:
|
| 636 |
-
value: null
|
| 637 |
-
torchdynamo:
|
| 638 |
-
value: null
|
| 639 |
-
torchscript:
|
| 640 |
-
value: false
|
| 641 |
-
tpu_metrics_debug:
|
| 642 |
-
value: false
|
| 643 |
-
tpu_num_cores:
|
| 644 |
-
value: null
|
| 645 |
-
trainer:
|
| 646 |
-
value:
|
| 647 |
-
action_dim: 26
|
| 648 |
-
dataset_text_field: text
|
| 649 |
-
max_seq_length: 100
|
| 650 |
-
obs_dim: 46
|
| 651 |
-
packing: false
|
| 652 |
-
use_joint_mlp_projector: true
|
| 653 |
-
training:
|
| 654 |
-
value:
|
| 655 |
-
bf16: true
|
| 656 |
-
eval_steps: 10
|
| 657 |
-
eval_strategy: steps
|
| 658 |
-
fp16: false
|
| 659 |
-
gradient_accumulation_steps: 1
|
| 660 |
-
group_by_length: true
|
| 661 |
-
joint_obs_action_mlp_lr: 5e-05
|
| 662 |
-
learning_rate: 0.0001
|
| 663 |
-
logging_steps: 1
|
| 664 |
-
logging_strategy: steps
|
| 665 |
-
num_train_epochs: 10
|
| 666 |
-
optim: paged_adamw_32bit
|
| 667 |
-
per_device_eval_batch_size: 128
|
| 668 |
-
per_device_train_batch_size: 128
|
| 669 |
-
report_to: wandb
|
| 670 |
-
save_steps: 5000
|
| 671 |
-
seed: 42
|
| 672 |
-
tf32: true
|
| 673 |
-
use_joint_mlp_projector: true
|
| 674 |
-
warmup_steps: 10
|
| 675 |
-
transformers.js_config:
|
| 676 |
-
value:
|
| 677 |
-
kv_cache_dtype:
|
| 678 |
-
fp16: float16
|
| 679 |
-
q4f16: float16
|
| 680 |
-
transformers_version:
|
| 681 |
-
value: 4.47.1
|
| 682 |
-
typical_p:
|
| 683 |
-
value: 1
|
| 684 |
-
use_bfloat16:
|
| 685 |
-
value: false
|
| 686 |
-
use_cache:
|
| 687 |
-
value: false
|
| 688 |
-
use_cpu:
|
| 689 |
-
value: false
|
| 690 |
-
use_ipex:
|
| 691 |
-
value: false
|
| 692 |
-
use_joint_mlp_projector:
|
| 693 |
-
value: true
|
| 694 |
-
use_legacy_prediction_loop:
|
| 695 |
-
value: false
|
| 696 |
-
use_liger:
|
| 697 |
-
value: false
|
| 698 |
-
use_liger_kernel:
|
| 699 |
-
value: false
|
| 700 |
-
use_mps_device:
|
| 701 |
-
value: false
|
| 702 |
-
use_quantization:
|
| 703 |
-
value: false
|
| 704 |
-
vocab_size:
|
| 705 |
-
value: 49152
|
| 706 |
-
warmup_ratio:
|
| 707 |
-
value: 0
|
| 708 |
-
warmup_steps:
|
| 709 |
-
value: 10
|
| 710 |
-
weight_decay:
|
| 711 |
-
value: 0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/files/output.log
DELETED
|
@@ -1,509 +0,0 @@
|
|
| 1 |
-
You are using a model of type llama to instantiate a model of type llama_lowdim. This is not supported for all configurations of models and can yield errors.
|
| 2 |
-
Some weights of LowdimLlamaForCausalLM were not initialized from the model checkpoint at HuggingFaceTB/SmolLM2-135M-Instruct and are newly initialized: ['model.joint_obs_action_projector.projector.0.bias', 'model.joint_obs_action_projector.projector.0.weight', 'model.joint_obs_action_projector.projector.2.bias', 'model.joint_obs_action_projector.projector.2.weight']
|
| 3 |
-
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
|
| 4 |
-
Loading from mlp projector checkpoint: data/outputs/2026.03.27/14.38.20_train_mlp_projector_adroit-hand-hammer-v1/checkpoints/latest.ckpt
|
| 5 |
-
Finetune the whole original LLM SmolLM2-135M-Instruct.
|
| 6 |
-
Multistep Flattening Dataset: 100%|██████████████████████████████████████████████████████████████████| 4585/4585 [00:00<00:00, 6904.74it/s]
|
| 7 |
-
Setting TOKENIZERS_PARALLELISM=false for forked processes.
|
| 8 |
-
[2026-03-27 16:20:59,889][datasets.arrow_dataset][WARNING] - Setting TOKENIZERS_PARALLELISM=false for forked processes.
|
| 9 |
-
Map (num_proc=4): 100%|███████████████████████████████████████████████████████████████████████| 4585/4585 [00:00<00:00, 4937.91 examples/s]
|
| 10 |
-
Setting TOKENIZERS_PARALLELISM=false for forked processes.
|
| 11 |
-
[2026-03-27 16:21:01,023][datasets.arrow_dataset][WARNING] - Setting TOKENIZERS_PARALLELISM=false for forked processes.
|
| 12 |
-
Map (num_proc=4): 100%|███████████████████████████████████████████████████████████████████████| 4585/4585 [00:00<00:00, 9277.34 examples/s]
|
| 13 |
-
DatasetDict({
|
| 14 |
-
train: Dataset({
|
| 15 |
-
features: ['obs', 'action', 'description', 'input', 'output', 'text', 'input_ids', 'labels'],
|
| 16 |
-
num_rows: 4539
|
| 17 |
-
})
|
| 18 |
-
test: Dataset({
|
| 19 |
-
features: ['obs', 'action', 'description', 'input', 'output', 'text', 'input_ids', 'labels'],
|
| 20 |
-
num_rows: 46
|
| 21 |
-
})
|
| 22 |
-
})
|
| 23 |
-
/home/chyang/miniconda3/envs/llm-bc/lib/python3.9/site-packages/huggingface_hub/utils/_deprecation.py:100: FutureWarning: Deprecated argument(s) used in '__init__': max_seq_length, dataset_text_field. Will not be supported from version '1.0.0'.
|
| 24 |
-
|
| 25 |
-
Deprecated positional argument(s) used in SFTTrainer, please use the SFTConfig to set these arguments instead.
|
| 26 |
-
warnings.warn(message, FutureWarning)
|
| 27 |
-
/home/chyang/miniconda3/envs/llm-bc/lib/python3.9/site-packages/trl/trainer/sft_trainer.py:283: UserWarning: You passed a `max_seq_length` argument to the SFTTrainer, the value you passed will override the one in the `SFTConfig`.
|
| 28 |
-
warnings.warn(
|
| 29 |
-
/home/chyang/miniconda3/envs/llm-bc/lib/python3.9/site-packages/trl/trainer/sft_trainer.py:321: UserWarning: You passed a `dataset_text_field` argument to the SFTTrainer, the value you passed will override the one in the `SFTConfig`.
|
| 30 |
-
warnings.warn(
|
| 31 |
-
/home/chyang/miniconda3/envs/llm-bc/lib/python3.9/site-packages/trl/trainer/sft_trainer.py:401: FutureWarning: `tokenizer` is deprecated and will be removed in version 5.0.0 for `LowdimSFTTrainer.__init__`. Use `processing_class` instead.
|
| 32 |
-
super().__init__(
|
| 33 |
-
[2026-03-27 16:21:01,783] [INFO] [real_accelerator.py:219:get_accelerator] Setting ds_accelerator to cuda (auto detect)
|
| 34 |
-
[2026-03-27 16:21:01,919][root][INFO] - gcc -pthread -B /home/chyang/miniconda3/envs/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /home/chyang/miniconda3/envs/llm-bc/include -I/home/chyang/miniconda3/envs/llm-bc/include -fPIC -O2 -isystem /home/chyang/miniconda3/envs/llm-bc/include -fPIC -c /tmp/tmp7t0nrkuh/test.c -o /tmp/tmp7t0nrkuh/test.o
|
| 35 |
-
[2026-03-27 16:21:01,969][root][INFO] - gcc -pthread -B /home/chyang/miniconda3/envs/llm-bc/compiler_compat /tmp/tmp7t0nrkuh/test.o -laio -o /tmp/tmp7t0nrkuh/a.out
|
| 36 |
-
[2026-03-27 16:21:02,963][root][INFO] - gcc -pthread -B /home/chyang/miniconda3/envs/llm-bc/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /home/chyang/miniconda3/envs/llm-bc/include -I/home/chyang/miniconda3/envs/llm-bc/include -fPIC -O2 -isystem /home/chyang/miniconda3/envs/llm-bc/include -fPIC -c /tmp/tmps_s5vyh9/test.c -o /tmp/tmps_s5vyh9/test.o
|
| 37 |
-
[2026-03-27 16:21:03,018][root][INFO] - gcc -pthread -B /home/chyang/miniconda3/envs/llm-bc/compiler_compat /tmp/tmps_s5vyh9/test.o -L/usr/local/cuda -L/usr/local/cuda/lib64 -lcufile -o /tmp/tmps_s5vyh9/a.out
|
| 38 |
-
[34m[1mwandb[0m: [33mWARNING[0m The `run_name` is currently set to the same value as `TrainingArguments.output_dir`. If this was not intended, please specify a different run name by setting the `TrainingArguments.run_name` parameter.
|
| 39 |
-
3%|██▊ | 10/360 [00:08<04:32, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 40 |
-
{'loss': 1.6431, 'grad_norm': 13.871646881103516, 'learning_rate': 1e-05, 'epoch': 0.03}
|
| 41 |
-
{'loss': 1.6018, 'grad_norm': 17.890308380126953, 'learning_rate': 2e-05, 'epoch': 0.06}
|
| 42 |
-
{'loss': 1.5953, 'grad_norm': 13.746294021606445, 'learning_rate': 3e-05, 'epoch': 0.08}
|
| 43 |
-
{'loss': 1.5355, 'grad_norm': 15.9970121383667, 'learning_rate': 4e-05, 'epoch': 0.11}
|
| 44 |
-
{'loss': 1.552, 'grad_norm': 18.634761810302734, 'learning_rate': 5e-05, 'epoch': 0.14}
|
| 45 |
-
{'loss': 1.4926, 'grad_norm': 10.22042179107666, 'learning_rate': 6e-05, 'epoch': 0.17}
|
| 46 |
-
{'loss': 1.1827, 'grad_norm': 14.976595878601074, 'learning_rate': 7e-05, 'epoch': 0.19}
|
| 47 |
-
{'loss': 1.173, 'grad_norm': 34.35334396362305, 'learning_rate': 8e-05, 'epoch': 0.22}
|
| 48 |
-
{'loss': 0.7987, 'grad_norm': 7.392702579498291, 'learning_rate': 9e-05, 'epoch': 0.25}
|
| 49 |
-
{'loss': 0.7641, 'grad_norm': 8.6481294631958, 'learning_rate': 0.0001, 'epoch': 0.28}
|
| 50 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 51 |
-
6%|█████▌ | 20/360 [00:15<04:21, 1.30it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 52 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 53 |
-
{'eval_loss': 0.719104528427124, 'eval_runtime': 0.1041, 'eval_samples_per_second': 441.967, 'eval_steps_per_second': 9.608, 'epoch': 0.28}
|
| 54 |
-
{'loss': 0.6656, 'grad_norm': 11.896777153015137, 'learning_rate': 9.971428571428571e-05, 'epoch': 0.31}
|
| 55 |
-
{'loss': 0.5743, 'grad_norm': 6.002552032470703, 'learning_rate': 9.942857142857144e-05, 'epoch': 0.33}
|
| 56 |
-
{'loss': 0.5204, 'grad_norm': 4.569210529327393, 'learning_rate': 9.914285714285715e-05, 'epoch': 0.36}
|
| 57 |
-
{'loss': 0.5217, 'grad_norm': 3.2792978286743164, 'learning_rate': 9.885714285714286e-05, 'epoch': 0.39}
|
| 58 |
-
{'loss': 0.5046, 'grad_norm': 3.6701369285583496, 'learning_rate': 9.857142857142858e-05, 'epoch': 0.42}
|
| 59 |
-
{'loss': 0.4803, 'grad_norm': 4.064358711242676, 'learning_rate': 9.828571428571429e-05, 'epoch': 0.44}
|
| 60 |
-
{'loss': 0.4466, 'grad_norm': 2.9059529304504395, 'learning_rate': 9.8e-05, 'epoch': 0.47}
|
| 61 |
-
{'loss': 0.4453, 'grad_norm': 2.499434471130371, 'learning_rate': 9.771428571428572e-05, 'epoch': 0.5}
|
| 62 |
-
{'loss': 0.4382, 'grad_norm': 2.084019899368286, 'learning_rate': 9.742857142857143e-05, 'epoch': 0.53}
|
| 63 |
-
{'loss': 0.4208, 'grad_norm': 1.2787052392959595, 'learning_rate': 9.714285714285715e-05, 'epoch': 0.56}
|
| 64 |
-
8%|████████▍ | 30/360 [00:23<04:10, 1.32it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 65 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 66 |
-
{'eval_loss': 0.4429771900177002, 'eval_runtime': 0.1036, 'eval_samples_per_second': 443.983, 'eval_steps_per_second': 9.652, 'epoch': 0.56}
|
| 67 |
-
{'loss': 0.431, 'grad_norm': 1.8704657554626465, 'learning_rate': 9.685714285714286e-05, 'epoch': 0.58}
|
| 68 |
-
{'loss': 0.4317, 'grad_norm': 1.7761015892028809, 'learning_rate': 9.657142857142858e-05, 'epoch': 0.61}
|
| 69 |
-
{'loss': 0.4229, 'grad_norm': 1.7499357461929321, 'learning_rate': 9.628571428571429e-05, 'epoch': 0.64}
|
| 70 |
-
{'loss': 0.4126, 'grad_norm': 1.2509069442749023, 'learning_rate': 9.6e-05, 'epoch': 0.67}
|
| 71 |
-
{'loss': 0.3888, 'grad_norm': 1.3415058851242065, 'learning_rate': 9.571428571428573e-05, 'epoch': 0.69}
|
| 72 |
-
{'loss': 0.4111, 'grad_norm': 1.513482689857483, 'learning_rate': 9.542857142857143e-05, 'epoch': 0.72}
|
| 73 |
-
{'loss': 0.3904, 'grad_norm': 1.0207685232162476, 'learning_rate': 9.514285714285714e-05, 'epoch': 0.75}
|
| 74 |
-
{'loss': 0.3911, 'grad_norm': 1.0765091180801392, 'learning_rate': 9.485714285714287e-05, 'epoch': 0.78}
|
| 75 |
-
{'loss': 0.3893, 'grad_norm': 1.2146029472351074, 'learning_rate': 9.457142857142858e-05, 'epoch': 0.81}
|
| 76 |
-
{'loss': 0.3915, 'grad_norm': 1.302972435951233, 'learning_rate': 9.428571428571429e-05, 'epoch': 0.83}
|
| 77 |
-
11%|███████████▏ | 40/360 [00:30<03:54, 1.36it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 78 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 79 |
-
{'eval_loss': 0.39956018328666687, 'eval_runtime': 0.1036, 'eval_samples_per_second': 444.089, 'eval_steps_per_second': 9.654, 'epoch': 0.83}
|
| 80 |
-
{'loss': 0.377, 'grad_norm': 1.2195433378219604, 'learning_rate': 9.4e-05, 'epoch': 0.86}
|
| 81 |
-
{'loss': 0.3837, 'grad_norm': 1.2320094108581543, 'learning_rate': 9.371428571428572e-05, 'epoch': 0.89}
|
| 82 |
-
{'loss': 0.3776, 'grad_norm': 1.0609043836593628, 'learning_rate': 9.342857142857143e-05, 'epoch': 0.92}
|
| 83 |
-
{'loss': 0.3887, 'grad_norm': 0.9609966278076172, 'learning_rate': 9.314285714285715e-05, 'epoch': 0.94}
|
| 84 |
-
{'loss': 0.3761, 'grad_norm': 1.0595581531524658, 'learning_rate': 9.285714285714286e-05, 'epoch': 0.97}
|
| 85 |
-
{'loss': 0.3744, 'grad_norm': 0.990327775478363, 'learning_rate': 9.257142857142858e-05, 'epoch': 1.0}
|
| 86 |
-
{'loss': 0.3625, 'grad_norm': 1.272873044013977, 'learning_rate': 9.228571428571429e-05, 'epoch': 1.03}
|
| 87 |
-
{'loss': 0.3869, 'grad_norm': 1.9024567604064941, 'learning_rate': 9.200000000000001e-05, 'epoch': 1.06}
|
| 88 |
-
{'loss': 0.3751, 'grad_norm': 1.3398654460906982, 'learning_rate': 9.171428571428572e-05, 'epoch': 1.08}
|
| 89 |
-
{'loss': 0.3662, 'grad_norm': 1.9176064729690552, 'learning_rate': 9.142857142857143e-05, 'epoch': 1.11}
|
| 90 |
-
14%|██████████████ | 50/360 [00:38<03:57, 1.31it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 91 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 92 |
-
{'eval_loss': 0.37765416502952576, 'eval_runtime': 0.1036, 'eval_samples_per_second': 444.001, 'eval_steps_per_second': 9.652, 'epoch': 1.11}
|
| 93 |
-
{'loss': 0.3665, 'grad_norm': 1.1852660179138184, 'learning_rate': 9.114285714285716e-05, 'epoch': 1.14}
|
| 94 |
-
{'loss': 0.3705, 'grad_norm': 1.831186056137085, 'learning_rate': 9.085714285714286e-05, 'epoch': 1.17}
|
| 95 |
-
{'loss': 0.3582, 'grad_norm': 1.1574777364730835, 'learning_rate': 9.057142857142857e-05, 'epoch': 1.19}
|
| 96 |
-
{'loss': 0.3724, 'grad_norm': 1.3485198020935059, 'learning_rate': 9.028571428571428e-05, 'epoch': 1.22}
|
| 97 |
-
{'loss': 0.3549, 'grad_norm': 1.0934721231460571, 'learning_rate': 9e-05, 'epoch': 1.25}
|
| 98 |
-
{'loss': 0.3607, 'grad_norm': 1.2588518857955933, 'learning_rate': 8.971428571428571e-05, 'epoch': 1.28}
|
| 99 |
-
{'loss': 0.3492, 'grad_norm': 0.9038533568382263, 'learning_rate': 8.942857142857142e-05, 'epoch': 1.31}
|
| 100 |
-
{'loss': 0.361, 'grad_norm': 1.083348274230957, 'learning_rate': 8.914285714285715e-05, 'epoch': 1.33}
|
| 101 |
-
{'loss': 0.3475, 'grad_norm': 0.8287424445152283, 'learning_rate': 8.885714285714286e-05, 'epoch': 1.36}
|
| 102 |
-
{'loss': 0.363, 'grad_norm': 1.3475714921951294, 'learning_rate': 8.857142857142857e-05, 'epoch': 1.39}
|
| 103 |
-
17%|████████████████▊ | 60/360 [00:46<03:47, 1.32it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 104 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 105 |
-
{'eval_loss': 0.3627206087112427, 'eval_runtime': 0.1039, 'eval_samples_per_second': 442.561, 'eval_steps_per_second': 9.621, 'epoch': 1.39}
|
| 106 |
-
{'loss': 0.3527, 'grad_norm': 1.1012552976608276, 'learning_rate': 8.828571428571429e-05, 'epoch': 1.42}
|
| 107 |
-
{'loss': 0.3418, 'grad_norm': 0.6421935558319092, 'learning_rate': 8.800000000000001e-05, 'epoch': 1.44}
|
| 108 |
-
{'loss': 0.3513, 'grad_norm': 1.1574995517730713, 'learning_rate': 8.771428571428572e-05, 'epoch': 1.47}
|
| 109 |
-
{'loss': 0.3515, 'grad_norm': 1.0251258611679077, 'learning_rate': 8.742857142857144e-05, 'epoch': 1.5}
|
| 110 |
-
{'loss': 0.3609, 'grad_norm': 0.9864039421081543, 'learning_rate': 8.714285714285715e-05, 'epoch': 1.53}
|
| 111 |
-
{'loss': 0.3454, 'grad_norm': 0.757999062538147, 'learning_rate': 8.685714285714286e-05, 'epoch': 1.56}
|
| 112 |
-
{'loss': 0.3488, 'grad_norm': 1.0983614921569824, 'learning_rate': 8.657142857142858e-05, 'epoch': 1.58}
|
| 113 |
-
{'loss': 0.3562, 'grad_norm': 1.4811136722564697, 'learning_rate': 8.62857142857143e-05, 'epoch': 1.61}
|
| 114 |
-
{'loss': 0.349, 'grad_norm': 0.9457672834396362, 'learning_rate': 8.6e-05, 'epoch': 1.64}
|
| 115 |
-
{'loss': 0.3551, 'grad_norm': 1.4347460269927979, 'learning_rate': 8.571428571428571e-05, 'epoch': 1.67}
|
| 116 |
-
19%|███████████████████▋ | 70/360 [00:54<03:43, 1.30it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 117 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 118 |
-
{'eval_loss': 0.35965070128440857, 'eval_runtime': 0.1045, 'eval_samples_per_second': 440.164, 'eval_steps_per_second': 9.569, 'epoch': 1.67}
|
| 119 |
-
{'loss': 0.3485, 'grad_norm': 1.0592706203460693, 'learning_rate': 8.542857142857144e-05, 'epoch': 1.69}
|
| 120 |
-
{'loss': 0.3512, 'grad_norm': 1.3444126844406128, 'learning_rate': 8.514285714285714e-05, 'epoch': 1.72}
|
| 121 |
-
{'loss': 0.3525, 'grad_norm': 0.9045667052268982, 'learning_rate': 8.485714285714285e-05, 'epoch': 1.75}
|
| 122 |
-
{'loss': 0.3483, 'grad_norm': 1.135429859161377, 'learning_rate': 8.457142857142858e-05, 'epoch': 1.78}
|
| 123 |
-
{'loss': 0.3445, 'grad_norm': 0.7742411494255066, 'learning_rate': 8.428571428571429e-05, 'epoch': 1.81}
|
| 124 |
-
{'loss': 0.3425, 'grad_norm': 1.2747840881347656, 'learning_rate': 8.4e-05, 'epoch': 1.83}
|
| 125 |
-
{'loss': 0.3506, 'grad_norm': 1.1280975341796875, 'learning_rate': 8.371428571428572e-05, 'epoch': 1.86}
|
| 126 |
-
{'loss': 0.3458, 'grad_norm': 1.3229925632476807, 'learning_rate': 8.342857142857143e-05, 'epoch': 1.89}
|
| 127 |
-
{'loss': 0.3443, 'grad_norm': 1.0970568656921387, 'learning_rate': 8.314285714285715e-05, 'epoch': 1.92}
|
| 128 |
-
{'loss': 0.3612, 'grad_norm': 1.7599389553070068, 'learning_rate': 8.285714285714287e-05, 'epoch': 1.94}
|
| 129 |
-
22%|██████████████████████▍ | 80/360 [01:01<03:32, 1.32it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 130 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 131 |
-
{'eval_loss': 0.35299769043922424, 'eval_runtime': 0.1042, 'eval_samples_per_second': 441.433, 'eval_steps_per_second': 9.596, 'epoch': 1.94}
|
| 132 |
-
{'loss': 0.3373, 'grad_norm': 0.8275991678237915, 'learning_rate': 8.257142857142858e-05, 'epoch': 1.97}
|
| 133 |
-
{'loss': 0.3624, 'grad_norm': 1.5045437812805176, 'learning_rate': 8.228571428571429e-05, 'epoch': 2.0}
|
| 134 |
-
{'loss': 0.3434, 'grad_norm': 0.9771829843521118, 'learning_rate': 8.2e-05, 'epoch': 2.03}
|
| 135 |
-
{'loss': 0.3347, 'grad_norm': 0.8552800416946411, 'learning_rate': 8.171428571428572e-05, 'epoch': 2.06}
|
| 136 |
-
{'loss': 0.3448, 'grad_norm': 0.8917291164398193, 'learning_rate': 8.142857142857143e-05, 'epoch': 2.08}
|
| 137 |
-
{'loss': 0.3309, 'grad_norm': 0.9143850207328796, 'learning_rate': 8.114285714285714e-05, 'epoch': 2.11}
|
| 138 |
-
{'loss': 0.3471, 'grad_norm': 1.359926700592041, 'learning_rate': 8.085714285714287e-05, 'epoch': 2.14}
|
| 139 |
-
{'loss': 0.3433, 'grad_norm': 0.84107506275177, 'learning_rate': 8.057142857142857e-05, 'epoch': 2.17}
|
| 140 |
-
{'loss': 0.3495, 'grad_norm': 1.2953639030456543, 'learning_rate': 8.028571428571428e-05, 'epoch': 2.19}
|
| 141 |
-
{'loss': 0.3388, 'grad_norm': 0.9937311410903931, 'learning_rate': 8e-05, 'epoch': 2.22}
|
| 142 |
-
25%|█████████████████████████▎ | 90/360 [01:09<03:26, 1.31it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 143 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 144 |
-
{'eval_loss': 0.3491460978984833, 'eval_runtime': 0.1044, 'eval_samples_per_second': 440.593, 'eval_steps_per_second': 9.578, 'epoch': 2.22}
|
| 145 |
-
{'loss': 0.3435, 'grad_norm': 1.0681490898132324, 'learning_rate': 7.971428571428572e-05, 'epoch': 2.25}
|
| 146 |
-
{'loss': 0.3333, 'grad_norm': 0.8466928601264954, 'learning_rate': 7.942857142857143e-05, 'epoch': 2.28}
|
| 147 |
-
{'loss': 0.3305, 'grad_norm': 0.8183342814445496, 'learning_rate': 7.914285714285715e-05, 'epoch': 2.31}
|
| 148 |
-
{'loss': 0.3289, 'grad_norm': 0.833314061164856, 'learning_rate': 7.885714285714286e-05, 'epoch': 2.33}
|
| 149 |
-
{'loss': 0.3331, 'grad_norm': 0.8347731828689575, 'learning_rate': 7.857142857142858e-05, 'epoch': 2.36}
|
| 150 |
-
{'loss': 0.3437, 'grad_norm': 1.0877679586410522, 'learning_rate': 7.828571428571429e-05, 'epoch': 2.39}
|
| 151 |
-
{'loss': 0.3331, 'grad_norm': 0.9570125937461853, 'learning_rate': 7.800000000000001e-05, 'epoch': 2.42}
|
| 152 |
-
{'loss': 0.3363, 'grad_norm': 0.7662280797958374, 'learning_rate': 7.771428571428572e-05, 'epoch': 2.44}
|
| 153 |
-
{'loss': 0.3305, 'grad_norm': 0.9321999549865723, 'learning_rate': 7.742857142857143e-05, 'epoch': 2.47}
|
| 154 |
-
{'loss': 0.3332, 'grad_norm': 0.8284544348716736, 'learning_rate': 7.714285714285715e-05, 'epoch': 2.5}
|
| 155 |
-
28%|███████████████████████████▊ | 100/360 [01:17<03:19, 1.30it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 156 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 157 |
-
{'eval_loss': 0.34498000144958496, 'eval_runtime': 0.1046, 'eval_samples_per_second': 439.848, 'eval_steps_per_second': 9.562, 'epoch': 2.5}
|
| 158 |
-
{'loss': 0.3448, 'grad_norm': 1.0568827390670776, 'learning_rate': 7.685714285714286e-05, 'epoch': 2.53}
|
| 159 |
-
{'loss': 0.334, 'grad_norm': 0.9136806130409241, 'learning_rate': 7.657142857142857e-05, 'epoch': 2.56}
|
| 160 |
-
{'loss': 0.3425, 'grad_norm': 1.2551990747451782, 'learning_rate': 7.62857142857143e-05, 'epoch': 2.58}
|
| 161 |
-
{'loss': 0.3265, 'grad_norm': 0.8284862637519836, 'learning_rate': 7.6e-05, 'epoch': 2.61}
|
| 162 |
-
{'loss': 0.3267, 'grad_norm': 0.7161554098129272, 'learning_rate': 7.571428571428571e-05, 'epoch': 2.64}
|
| 163 |
-
{'loss': 0.3342, 'grad_norm': 0.8050905466079712, 'learning_rate': 7.542857142857144e-05, 'epoch': 2.67}
|
| 164 |
-
{'loss': 0.3325, 'grad_norm': 0.7441209554672241, 'learning_rate': 7.514285714285715e-05, 'epoch': 2.69}
|
| 165 |
-
{'loss': 0.334, 'grad_norm': 0.591927707195282, 'learning_rate': 7.485714285714285e-05, 'epoch': 2.72}
|
| 166 |
-
{'loss': 0.3473, 'grad_norm': 0.8902866244316101, 'learning_rate': 7.457142857142856e-05, 'epoch': 2.75}
|
| 167 |
-
{'loss': 0.326, 'grad_norm': 0.6760069131851196, 'learning_rate': 7.428571428571429e-05, 'epoch': 2.78}
|
| 168 |
-
31%|██████████████████████████████▌ | 110/360 [01:24<02:57, 1.41it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 169 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 170 |
-
{'eval_loss': 0.3410107493400574, 'eval_runtime': 0.1043, 'eval_samples_per_second': 441.026, 'eval_steps_per_second': 9.588, 'epoch': 2.78}
|
| 171 |
-
{'loss': 0.3348, 'grad_norm': 0.6579345464706421, 'learning_rate': 7.4e-05, 'epoch': 2.81}
|
| 172 |
-
{'loss': 0.3253, 'grad_norm': 1.0648415088653564, 'learning_rate': 7.371428571428572e-05, 'epoch': 2.83}
|
| 173 |
-
{'loss': 0.3297, 'grad_norm': 0.6868814826011658, 'learning_rate': 7.342857142857144e-05, 'epoch': 2.86}
|
| 174 |
-
{'loss': 0.3401, 'grad_norm': 1.1149464845657349, 'learning_rate': 7.314285714285715e-05, 'epoch': 2.89}
|
| 175 |
-
{'loss': 0.3348, 'grad_norm': 0.8934164047241211, 'learning_rate': 7.285714285714286e-05, 'epoch': 2.92}
|
| 176 |
-
{'loss': 0.3427, 'grad_norm': 1.1119507551193237, 'learning_rate': 7.257142857142858e-05, 'epoch': 2.94}
|
| 177 |
-
{'loss': 0.3374, 'grad_norm': 0.8103634715080261, 'learning_rate': 7.228571428571429e-05, 'epoch': 2.97}
|
| 178 |
-
{'loss': 0.3395, 'grad_norm': 0.8421126008033752, 'learning_rate': 7.2e-05, 'epoch': 3.0}
|
| 179 |
-
{'loss': 0.331, 'grad_norm': 0.8583278656005859, 'learning_rate': 7.171428571428572e-05, 'epoch': 3.03}
|
| 180 |
-
{'loss': 0.3355, 'grad_norm': 1.2129111289978027, 'learning_rate': 7.142857142857143e-05, 'epoch': 3.06}
|
| 181 |
-
33%|█████████████████████████████████▎ | 120/360 [01:32<03:02, 1.32it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 182 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 183 |
-
{'eval_loss': 0.332057386636734, 'eval_runtime': 0.1051, 'eval_samples_per_second': 437.685, 'eval_steps_per_second': 9.515, 'epoch': 3.06}
|
| 184 |
-
{'loss': 0.3294, 'grad_norm': 0.9463130235671997, 'learning_rate': 7.114285714285714e-05, 'epoch': 3.08}
|
| 185 |
-
{'loss': 0.3327, 'grad_norm': 0.9692079424858093, 'learning_rate': 7.085714285714285e-05, 'epoch': 3.11}
|
| 186 |
-
{'loss': 0.3295, 'grad_norm': 0.9853659868240356, 'learning_rate': 7.057142857142858e-05, 'epoch': 3.14}
|
| 187 |
-
{'loss': 0.3358, 'grad_norm': 0.7222715616226196, 'learning_rate': 7.028571428571428e-05, 'epoch': 3.17}
|
| 188 |
-
{'loss': 0.3406, 'grad_norm': 1.1528452634811401, 'learning_rate': 7e-05, 'epoch': 3.19}
|
| 189 |
-
{'loss': 0.329, 'grad_norm': 1.0079970359802246, 'learning_rate': 6.971428571428572e-05, 'epoch': 3.22}
|
| 190 |
-
{'loss': 0.327, 'grad_norm': 0.7162885665893555, 'learning_rate': 6.942857142857143e-05, 'epoch': 3.25}
|
| 191 |
-
{'loss': 0.336, 'grad_norm': 0.9302375912666321, 'learning_rate': 6.914285714285715e-05, 'epoch': 3.28}
|
| 192 |
-
{'loss': 0.3356, 'grad_norm': 0.8540468215942383, 'learning_rate': 6.885714285714286e-05, 'epoch': 3.31}
|
| 193 |
-
{'loss': 0.3279, 'grad_norm': 0.598040759563446, 'learning_rate': 6.857142857142858e-05, 'epoch': 3.33}
|
| 194 |
-
36%|████████████████████████████████████ | 130/360 [01:40<02:57, 1.30it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 195 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 196 |
-
{'eval_loss': 0.3290035128593445, 'eval_runtime': 0.1049, 'eval_samples_per_second': 438.622, 'eval_steps_per_second': 9.535, 'epoch': 3.33}
|
| 197 |
-
{'loss': 0.33, 'grad_norm': 0.6981043815612793, 'learning_rate': 6.828571428571429e-05, 'epoch': 3.36}
|
| 198 |
-
{'loss': 0.3262, 'grad_norm': 0.6710860133171082, 'learning_rate': 6.800000000000001e-05, 'epoch': 3.39}
|
| 199 |
-
{'loss': 0.3273, 'grad_norm': 0.6621596813201904, 'learning_rate': 6.771428571428572e-05, 'epoch': 3.42}
|
| 200 |
-
{'loss': 0.3264, 'grad_norm': 0.8255563974380493, 'learning_rate': 6.742857142857143e-05, 'epoch': 3.44}
|
| 201 |
-
{'loss': 0.3157, 'grad_norm': 0.8350001573562622, 'learning_rate': 6.714285714285714e-05, 'epoch': 3.47}
|
| 202 |
-
{'loss': 0.3205, 'grad_norm': 0.7275986075401306, 'learning_rate': 6.685714285714286e-05, 'epoch': 3.5}
|
| 203 |
-
{'loss': 0.3327, 'grad_norm': 0.652642548084259, 'learning_rate': 6.657142857142857e-05, 'epoch': 3.53}
|
| 204 |
-
{'loss': 0.3319, 'grad_norm': 0.9522960186004639, 'learning_rate': 6.628571428571428e-05, 'epoch': 3.56}
|
| 205 |
-
{'loss': 0.3292, 'grad_norm': 0.7006963491439819, 'learning_rate': 6.6e-05, 'epoch': 3.58}
|
| 206 |
-
{'loss': 0.3246, 'grad_norm': 0.7161970138549805, 'learning_rate': 6.571428571428571e-05, 'epoch': 3.61}
|
| 207 |
-
39%|██████████████████████████████████████▉ | 140/360 [01:48<02:51, 1.28it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 208 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 209 |
-
{'eval_loss': 0.3442412316799164, 'eval_runtime': 0.1052, 'eval_samples_per_second': 437.452, 'eval_steps_per_second': 9.51, 'epoch': 3.61}
|
| 210 |
-
{'loss': 0.3289, 'grad_norm': 1.0642709732055664, 'learning_rate': 6.542857142857142e-05, 'epoch': 3.64}
|
| 211 |
-
{'loss': 0.3234, 'grad_norm': 0.7999193072319031, 'learning_rate': 6.514285714285715e-05, 'epoch': 3.67}
|
| 212 |
-
{'loss': 0.3297, 'grad_norm': 0.8324876427650452, 'learning_rate': 6.485714285714286e-05, 'epoch': 3.69}
|
| 213 |
-
{'loss': 0.3125, 'grad_norm': 0.561801552772522, 'learning_rate': 6.457142857142856e-05, 'epoch': 3.72}
|
| 214 |
-
{'loss': 0.3234, 'grad_norm': 0.6995918154716492, 'learning_rate': 6.428571428571429e-05, 'epoch': 3.75}
|
| 215 |
-
{'loss': 0.3256, 'grad_norm': 0.6314477920532227, 'learning_rate': 6.400000000000001e-05, 'epoch': 3.78}
|
| 216 |
-
{'loss': 0.3315, 'grad_norm': 0.9092559814453125, 'learning_rate': 6.371428571428572e-05, 'epoch': 3.81}
|
| 217 |
-
{'loss': 0.3241, 'grad_norm': 0.7306588292121887, 'learning_rate': 6.342857142857143e-05, 'epoch': 3.83}
|
| 218 |
-
{'loss': 0.3323, 'grad_norm': 0.7943991422653198, 'learning_rate': 6.314285714285715e-05, 'epoch': 3.86}
|
| 219 |
-
{'loss': 0.3273, 'grad_norm': 0.8375313878059387, 'learning_rate': 6.285714285714286e-05, 'epoch': 3.89}
|
| 220 |
-
42%|█████████████████████████████████████████▋ | 150/360 [01:55<02:38, 1.33it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 221 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 222 |
-
{'eval_loss': 0.33936312794685364, 'eval_runtime': 0.1048, 'eval_samples_per_second': 438.851, 'eval_steps_per_second': 9.54, 'epoch': 3.89}
|
| 223 |
-
{'loss': 0.3258, 'grad_norm': 0.9479944705963135, 'learning_rate': 6.257142857142857e-05, 'epoch': 3.92}
|
| 224 |
-
{'loss': 0.3228, 'grad_norm': 0.8155922889709473, 'learning_rate': 6.22857142857143e-05, 'epoch': 3.94}
|
| 225 |
-
{'loss': 0.3217, 'grad_norm': 0.8617050647735596, 'learning_rate': 6.2e-05, 'epoch': 3.97}
|
| 226 |
-
{'loss': 0.3309, 'grad_norm': 1.2106715440750122, 'learning_rate': 6.171428571428571e-05, 'epoch': 4.0}
|
| 227 |
-
{'loss': 0.3265, 'grad_norm': 0.8097350001335144, 'learning_rate': 6.142857142857143e-05, 'epoch': 4.03}
|
| 228 |
-
{'loss': 0.3222, 'grad_norm': 0.651019811630249, 'learning_rate': 6.114285714285714e-05, 'epoch': 4.06}
|
| 229 |
-
{'loss': 0.3245, 'grad_norm': 0.8858047127723694, 'learning_rate': 6.085714285714286e-05, 'epoch': 4.08}
|
| 230 |
-
{'loss': 0.3153, 'grad_norm': 0.8602396845817566, 'learning_rate': 6.0571428571428576e-05, 'epoch': 4.11}
|
| 231 |
-
{'loss': 0.313, 'grad_norm': 0.615274965763092, 'learning_rate': 6.028571428571429e-05, 'epoch': 4.14}
|
| 232 |
-
{'loss': 0.3275, 'grad_norm': 0.9199692010879517, 'learning_rate': 6e-05, 'epoch': 4.17}
|
| 233 |
-
44%|████████████████████████████████████████████▍ | 160/360 [02:03<02:34, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 234 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 235 |
-
{'eval_loss': 0.34145644307136536, 'eval_runtime': 0.1056, 'eval_samples_per_second': 435.469, 'eval_steps_per_second': 9.467, 'epoch': 4.17}
|
| 236 |
-
{'loss': 0.3287, 'grad_norm': 0.9348899722099304, 'learning_rate': 5.9714285714285724e-05, 'epoch': 4.19}
|
| 237 |
-
{'loss': 0.3203, 'grad_norm': 0.7250774502754211, 'learning_rate': 5.9428571428571434e-05, 'epoch': 4.22}
|
| 238 |
-
{'loss': 0.3169, 'grad_norm': 0.7376280426979065, 'learning_rate': 5.914285714285714e-05, 'epoch': 4.25}
|
| 239 |
-
{'loss': 0.3215, 'grad_norm': 0.6010245680809021, 'learning_rate': 5.885714285714285e-05, 'epoch': 4.28}
|
| 240 |
-
{'loss': 0.317, 'grad_norm': 0.7241640686988831, 'learning_rate': 5.8571428571428575e-05, 'epoch': 4.31}
|
| 241 |
-
{'loss': 0.3217, 'grad_norm': 0.6956952810287476, 'learning_rate': 5.828571428571429e-05, 'epoch': 4.33}
|
| 242 |
-
{'loss': 0.322, 'grad_norm': 0.8463672995567322, 'learning_rate': 5.8e-05, 'epoch': 4.36}
|
| 243 |
-
{'loss': 0.3129, 'grad_norm': 0.5538536906242371, 'learning_rate': 5.771428571428572e-05, 'epoch': 4.39}
|
| 244 |
-
{'loss': 0.3275, 'grad_norm': 0.8398566246032715, 'learning_rate': 5.742857142857143e-05, 'epoch': 4.42}
|
| 245 |
-
{'loss': 0.3225, 'grad_norm': 0.5335714221000671, 'learning_rate': 5.714285714285714e-05, 'epoch': 4.44}
|
| 246 |
-
47%|███████████████████████████████████████████████▏ | 170/360 [02:11<02:27, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 247 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 248 |
-
{'eval_loss': 0.3419412672519684, 'eval_runtime': 0.1054, 'eval_samples_per_second': 436.373, 'eval_steps_per_second': 9.486, 'epoch': 4.44}
|
| 249 |
-
{'loss': 0.3201, 'grad_norm': 0.893516480922699, 'learning_rate': 5.6857142857142865e-05, 'epoch': 4.47}
|
| 250 |
-
{'loss': 0.3246, 'grad_norm': 0.7950851321220398, 'learning_rate': 5.6571428571428574e-05, 'epoch': 4.5}
|
| 251 |
-
{'loss': 0.3138, 'grad_norm': 0.6274649500846863, 'learning_rate': 5.628571428571428e-05, 'epoch': 4.53}
|
| 252 |
-
{'loss': 0.3225, 'grad_norm': 0.6270793676376343, 'learning_rate': 5.6000000000000006e-05, 'epoch': 4.56}
|
| 253 |
-
{'loss': 0.3222, 'grad_norm': 0.6661616563796997, 'learning_rate': 5.571428571428572e-05, 'epoch': 4.58}
|
| 254 |
-
{'loss': 0.3138, 'grad_norm': 0.6097862124443054, 'learning_rate': 5.542857142857143e-05, 'epoch': 4.61}
|
| 255 |
-
{'loss': 0.3109, 'grad_norm': 0.6743194460868835, 'learning_rate': 5.514285714285714e-05, 'epoch': 4.64}
|
| 256 |
-
{'loss': 0.3193, 'grad_norm': 0.6684880256652832, 'learning_rate': 5.485714285714286e-05, 'epoch': 4.67}
|
| 257 |
-
{'loss': 0.3212, 'grad_norm': 0.7434603571891785, 'learning_rate': 5.457142857142857e-05, 'epoch': 4.69}
|
| 258 |
-
{'loss': 0.323, 'grad_norm': 0.8257206082344055, 'learning_rate': 5.428571428571428e-05, 'epoch': 4.72}
|
| 259 |
-
50%|██████████████████████████████████████████████████ | 180/360 [02:18<01:58, 1.52it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 260 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 261 |
-
{'eval_loss': 0.335285484790802, 'eval_runtime': 0.1056, 'eval_samples_per_second': 435.635, 'eval_steps_per_second': 9.47, 'epoch': 4.72}
|
| 262 |
-
{'loss': 0.3249, 'grad_norm': 0.5611714720726013, 'learning_rate': 5.4000000000000005e-05, 'epoch': 4.75}
|
| 263 |
-
{'loss': 0.3213, 'grad_norm': 0.6146838068962097, 'learning_rate': 5.3714285714285714e-05, 'epoch': 4.78}
|
| 264 |
-
{'loss': 0.3239, 'grad_norm': 0.6939036846160889, 'learning_rate': 5.342857142857143e-05, 'epoch': 4.81}
|
| 265 |
-
{'loss': 0.3234, 'grad_norm': 0.7213876843452454, 'learning_rate': 5.314285714285715e-05, 'epoch': 4.83}
|
| 266 |
-
{'loss': 0.3186, 'grad_norm': 0.6637408137321472, 'learning_rate': 5.285714285714286e-05, 'epoch': 4.86}
|
| 267 |
-
{'loss': 0.3184, 'grad_norm': 0.6469508409500122, 'learning_rate': 5.257142857142857e-05, 'epoch': 4.89}
|
| 268 |
-
{'loss': 0.3171, 'grad_norm': 0.6262702941894531, 'learning_rate': 5.2285714285714294e-05, 'epoch': 4.92}
|
| 269 |
-
{'loss': 0.3271, 'grad_norm': 0.6692309379577637, 'learning_rate': 5.2000000000000004e-05, 'epoch': 4.94}
|
| 270 |
-
{'loss': 0.3229, 'grad_norm': 0.611004114151001, 'learning_rate': 5.171428571428571e-05, 'epoch': 4.97}
|
| 271 |
-
{'loss': 0.3223, 'grad_norm': 0.9707463383674622, 'learning_rate': 5.142857142857143e-05, 'epoch': 5.0}
|
| 272 |
-
53%|████████████████████████████████████████████████████▊ | 190/360 [02:26<02:11, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 273 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 274 |
-
{'eval_loss': 0.33622071146965027, 'eval_runtime': 0.105, 'eval_samples_per_second': 437.997, 'eval_steps_per_second': 9.522, 'epoch': 5.0}
|
| 275 |
-
{'loss': 0.3106, 'grad_norm': 0.43101295828819275, 'learning_rate': 5.1142857142857145e-05, 'epoch': 5.03}
|
| 276 |
-
{'loss': 0.3139, 'grad_norm': 0.7981957793235779, 'learning_rate': 5.085714285714286e-05, 'epoch': 5.06}
|
| 277 |
-
{'loss': 0.3137, 'grad_norm': 0.9149967432022095, 'learning_rate': 5.057142857142857e-05, 'epoch': 5.08}
|
| 278 |
-
{'loss': 0.3185, 'grad_norm': 0.8689376711845398, 'learning_rate': 5.028571428571429e-05, 'epoch': 5.11}
|
| 279 |
-
{'loss': 0.3195, 'grad_norm': 0.6829914450645447, 'learning_rate': 5e-05, 'epoch': 5.14}
|
| 280 |
-
{'loss': 0.3139, 'grad_norm': 0.6187098026275635, 'learning_rate': 4.971428571428572e-05, 'epoch': 5.17}
|
| 281 |
-
{'loss': 0.3147, 'grad_norm': 0.8703141212463379, 'learning_rate': 4.942857142857143e-05, 'epoch': 5.19}
|
| 282 |
-
{'loss': 0.3162, 'grad_norm': 0.6344360709190369, 'learning_rate': 4.9142857142857144e-05, 'epoch': 5.22}
|
| 283 |
-
{'loss': 0.3228, 'grad_norm': 0.7499691843986511, 'learning_rate': 4.885714285714286e-05, 'epoch': 5.25}
|
| 284 |
-
{'loss': 0.3152, 'grad_norm': 0.7664843201637268, 'learning_rate': 4.8571428571428576e-05, 'epoch': 5.28}
|
| 285 |
-
56%|███████████████████████████████████████████████████████▌ | 200/360 [02:34<02:03, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 286 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 287 |
-
{'eval_loss': 0.3364305794239044, 'eval_runtime': 0.1056, 'eval_samples_per_second': 435.778, 'eval_steps_per_second': 9.473, 'epoch': 5.28}
|
| 288 |
-
{'loss': 0.3154, 'grad_norm': 0.6158504486083984, 'learning_rate': 4.828571428571429e-05, 'epoch': 5.31}
|
| 289 |
-
{'loss': 0.3206, 'grad_norm': 0.8614490032196045, 'learning_rate': 4.8e-05, 'epoch': 5.33}
|
| 290 |
-
{'loss': 0.3159, 'grad_norm': 0.7699540257453918, 'learning_rate': 4.771428571428572e-05, 'epoch': 5.36}
|
| 291 |
-
{'loss': 0.3186, 'grad_norm': 0.9598901867866516, 'learning_rate': 4.742857142857143e-05, 'epoch': 5.39}
|
| 292 |
-
{'loss': 0.3118, 'grad_norm': 0.855253279209137, 'learning_rate': 4.714285714285714e-05, 'epoch': 5.42}
|
| 293 |
-
{'loss': 0.3178, 'grad_norm': 0.6478847861289978, 'learning_rate': 4.685714285714286e-05, 'epoch': 5.44}
|
| 294 |
-
{'loss': 0.3236, 'grad_norm': 0.8028067946434021, 'learning_rate': 4.6571428571428575e-05, 'epoch': 5.47}
|
| 295 |
-
{'loss': 0.3147, 'grad_norm': 0.7795782089233398, 'learning_rate': 4.628571428571429e-05, 'epoch': 5.5}
|
| 296 |
-
{'loss': 0.3221, 'grad_norm': 0.7845653891563416, 'learning_rate': 4.600000000000001e-05, 'epoch': 5.53}
|
| 297 |
-
{'loss': 0.321, 'grad_norm': 1.1422370672225952, 'learning_rate': 4.5714285714285716e-05, 'epoch': 5.56}
|
| 298 |
-
58%|██████████████████████████████████████████████████████████▎ | 210/360 [02:42<01:55, 1.30it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 299 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 300 |
-
{'eval_loss': 0.33535492420196533, 'eval_runtime': 0.1051, 'eval_samples_per_second': 437.57, 'eval_steps_per_second': 9.512, 'epoch': 5.56}
|
| 301 |
-
{'loss': 0.3183, 'grad_norm': 0.7386415600776672, 'learning_rate': 4.542857142857143e-05, 'epoch': 5.58}
|
| 302 |
-
{'loss': 0.3205, 'grad_norm': 0.6756716966629028, 'learning_rate': 4.514285714285714e-05, 'epoch': 5.61}
|
| 303 |
-
{'loss': 0.3195, 'grad_norm': 0.7116839289665222, 'learning_rate': 4.485714285714286e-05, 'epoch': 5.64}
|
| 304 |
-
{'loss': 0.3248, 'grad_norm': 0.7919530272483826, 'learning_rate': 4.4571428571428574e-05, 'epoch': 5.67}
|
| 305 |
-
{'loss': 0.3124, 'grad_norm': 0.5152342319488525, 'learning_rate': 4.428571428571428e-05, 'epoch': 5.69}
|
| 306 |
-
{'loss': 0.3205, 'grad_norm': 0.9519732594490051, 'learning_rate': 4.4000000000000006e-05, 'epoch': 5.72}
|
| 307 |
-
{'loss': 0.3188, 'grad_norm': 0.7759018540382385, 'learning_rate': 4.371428571428572e-05, 'epoch': 5.75}
|
| 308 |
-
{'loss': 0.3195, 'grad_norm': 0.931468665599823, 'learning_rate': 4.342857142857143e-05, 'epoch': 5.78}
|
| 309 |
-
{'loss': 0.3256, 'grad_norm': 1.0431036949157715, 'learning_rate': 4.314285714285715e-05, 'epoch': 5.81}
|
| 310 |
-
{'loss': 0.3189, 'grad_norm': 0.6952974200248718, 'learning_rate': 4.2857142857142856e-05, 'epoch': 5.83}
|
| 311 |
-
61%|█████████████████████████████████████████████████████████████ | 220/360 [02:49<01:45, 1.33it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 312 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 313 |
-
{'eval_loss': 0.33737656474113464, 'eval_runtime': 0.1056, 'eval_samples_per_second': 435.436, 'eval_steps_per_second': 9.466, 'epoch': 5.83}
|
| 314 |
-
{'loss': 0.3159, 'grad_norm': 0.880944013595581, 'learning_rate': 4.257142857142857e-05, 'epoch': 5.86}
|
| 315 |
-
{'loss': 0.3177, 'grad_norm': 0.6606886982917786, 'learning_rate': 4.228571428571429e-05, 'epoch': 5.89}
|
| 316 |
-
{'loss': 0.317, 'grad_norm': 0.7617785930633545, 'learning_rate': 4.2e-05, 'epoch': 5.92}
|
| 317 |
-
{'loss': 0.313, 'grad_norm': 0.6381199955940247, 'learning_rate': 4.1714285714285714e-05, 'epoch': 5.94}
|
| 318 |
-
{'loss': 0.3184, 'grad_norm': 0.6644593477249146, 'learning_rate': 4.1428571428571437e-05, 'epoch': 5.97}
|
| 319 |
-
{'loss': 0.3103, 'grad_norm': 0.9742204546928406, 'learning_rate': 4.1142857142857146e-05, 'epoch': 6.0}
|
| 320 |
-
{'loss': 0.311, 'grad_norm': 0.6401204466819763, 'learning_rate': 4.085714285714286e-05, 'epoch': 6.03}
|
| 321 |
-
{'loss': 0.3155, 'grad_norm': 0.785503089427948, 'learning_rate': 4.057142857142857e-05, 'epoch': 6.06}
|
| 322 |
-
{'loss': 0.3109, 'grad_norm': 0.5417785048484802, 'learning_rate': 4.028571428571429e-05, 'epoch': 6.08}
|
| 323 |
-
{'loss': 0.3114, 'grad_norm': 0.6186631321907043, 'learning_rate': 4e-05, 'epoch': 6.11}
|
| 324 |
-
64%|███████████████████████████████████████████████████████████████▉ | 230/360 [02:57<01:40, 1.30it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 325 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 326 |
-
{'eval_loss': 0.32762280106544495, 'eval_runtime': 0.1058, 'eval_samples_per_second': 434.862, 'eval_steps_per_second': 9.454, 'epoch': 6.11}
|
| 327 |
-
{'loss': 0.3099, 'grad_norm': 0.6042884588241577, 'learning_rate': 3.971428571428571e-05, 'epoch': 6.14}
|
| 328 |
-
{'loss': 0.3104, 'grad_norm': 0.5823580026626587, 'learning_rate': 3.942857142857143e-05, 'epoch': 6.17}
|
| 329 |
-
{'loss': 0.3154, 'grad_norm': 0.6572258472442627, 'learning_rate': 3.9142857142857145e-05, 'epoch': 6.19}
|
| 330 |
-
{'loss': 0.3121, 'grad_norm': 0.565834641456604, 'learning_rate': 3.885714285714286e-05, 'epoch': 6.22}
|
| 331 |
-
{'loss': 0.3119, 'grad_norm': 0.7184673547744751, 'learning_rate': 3.857142857142858e-05, 'epoch': 6.25}
|
| 332 |
-
{'loss': 0.3158, 'grad_norm': 0.7170347571372986, 'learning_rate': 3.8285714285714286e-05, 'epoch': 6.28}
|
| 333 |
-
{'loss': 0.3038, 'grad_norm': 0.6102560758590698, 'learning_rate': 3.8e-05, 'epoch': 6.31}
|
| 334 |
-
{'loss': 0.3124, 'grad_norm': 0.7612823843955994, 'learning_rate': 3.771428571428572e-05, 'epoch': 6.33}
|
| 335 |
-
{'loss': 0.3028, 'grad_norm': 0.6277872920036316, 'learning_rate': 3.742857142857143e-05, 'epoch': 6.36}
|
| 336 |
-
{'loss': 0.3202, 'grad_norm': 0.7007192373275757, 'learning_rate': 3.7142857142857143e-05, 'epoch': 6.39}
|
| 337 |
-
67%|██████████████████████████████████████████████████████████████████▋ | 240/360 [03:05<01:33, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 338 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 339 |
-
{'eval_loss': 0.33139991760253906, 'eval_runtime': 0.1062, 'eval_samples_per_second': 433.311, 'eval_steps_per_second': 9.42, 'epoch': 6.39}
|
| 340 |
-
{'loss': 0.313, 'grad_norm': 0.6396629810333252, 'learning_rate': 3.685714285714286e-05, 'epoch': 6.42}
|
| 341 |
-
{'loss': 0.3116, 'grad_norm': 0.5031012892723083, 'learning_rate': 3.6571428571428576e-05, 'epoch': 6.44}
|
| 342 |
-
{'loss': 0.3172, 'grad_norm': 0.7323219776153564, 'learning_rate': 3.628571428571429e-05, 'epoch': 6.47}
|
| 343 |
-
{'loss': 0.3103, 'grad_norm': 0.9094661474227905, 'learning_rate': 3.6e-05, 'epoch': 6.5}
|
| 344 |
-
{'loss': 0.3056, 'grad_norm': 0.5560885667800903, 'learning_rate': 3.571428571428572e-05, 'epoch': 6.53}
|
| 345 |
-
{'loss': 0.3096, 'grad_norm': 1.0145907402038574, 'learning_rate': 3.5428571428571426e-05, 'epoch': 6.56}
|
| 346 |
-
{'loss': 0.3049, 'grad_norm': 0.8287002444267273, 'learning_rate': 3.514285714285714e-05, 'epoch': 6.58}
|
| 347 |
-
{'loss': 0.3047, 'grad_norm': 0.5207920074462891, 'learning_rate': 3.485714285714286e-05, 'epoch': 6.61}
|
| 348 |
-
{'loss': 0.3156, 'grad_norm': 1.065272331237793, 'learning_rate': 3.4571428571428574e-05, 'epoch': 6.64}
|
| 349 |
-
{'loss': 0.3055, 'grad_norm': 0.6712301969528198, 'learning_rate': 3.428571428571429e-05, 'epoch': 6.67}
|
| 350 |
-
69%|█████████████████████████████████████████████████████████████████████▍ | 250/360 [03:13<01:25, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 351 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 352 |
-
{'eval_loss': 0.3387901484966278, 'eval_runtime': 0.1053, 'eval_samples_per_second': 436.76, 'eval_steps_per_second': 9.495, 'epoch': 6.67}
|
| 353 |
-
{'loss': 0.3164, 'grad_norm': 0.9477025866508484, 'learning_rate': 3.4000000000000007e-05, 'epoch': 6.69}
|
| 354 |
-
{'loss': 0.3263, 'grad_norm': 1.179295301437378, 'learning_rate': 3.3714285714285716e-05, 'epoch': 6.72}
|
| 355 |
-
{'loss': 0.3136, 'grad_norm': 0.6969395875930786, 'learning_rate': 3.342857142857143e-05, 'epoch': 6.75}
|
| 356 |
-
{'loss': 0.3143, 'grad_norm': 0.7752478718757629, 'learning_rate': 3.314285714285714e-05, 'epoch': 6.78}
|
| 357 |
-
{'loss': 0.3143, 'grad_norm': 0.7640174031257629, 'learning_rate': 3.285714285714286e-05, 'epoch': 6.81}
|
| 358 |
-
{'loss': 0.3093, 'grad_norm': 0.9904161691665649, 'learning_rate': 3.257142857142857e-05, 'epoch': 6.83}
|
| 359 |
-
{'loss': 0.3185, 'grad_norm': 0.7707158923149109, 'learning_rate': 3.228571428571428e-05, 'epoch': 6.86}
|
| 360 |
-
{'loss': 0.3122, 'grad_norm': 0.8660508394241333, 'learning_rate': 3.2000000000000005e-05, 'epoch': 6.89}
|
| 361 |
-
{'loss': 0.3148, 'grad_norm': 0.7438889741897583, 'learning_rate': 3.1714285714285715e-05, 'epoch': 6.92}
|
| 362 |
-
{'loss': 0.3137, 'grad_norm': 0.5746331810951233, 'learning_rate': 3.142857142857143e-05, 'epoch': 6.94}
|
| 363 |
-
72%|████████████████████████████████████████████████████████████████████████▏ | 260/360 [03:20<01:17, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 364 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 365 |
-
{'eval_loss': 0.32808226346969604, 'eval_runtime': 0.1059, 'eval_samples_per_second': 434.438, 'eval_steps_per_second': 9.444, 'epoch': 6.94}
|
| 366 |
-
{'loss': 0.3125, 'grad_norm': 0.7158817052841187, 'learning_rate': 3.114285714285715e-05, 'epoch': 6.97}
|
| 367 |
-
{'loss': 0.3094, 'grad_norm': 0.8010092973709106, 'learning_rate': 3.0857142857142856e-05, 'epoch': 7.0}
|
| 368 |
-
{'loss': 0.3067, 'grad_norm': 0.7418866157531738, 'learning_rate': 3.057142857142857e-05, 'epoch': 7.03}
|
| 369 |
-
{'loss': 0.3068, 'grad_norm': 0.6731083989143372, 'learning_rate': 3.0285714285714288e-05, 'epoch': 7.06}
|
| 370 |
-
{'loss': 0.3075, 'grad_norm': 0.6405408382415771, 'learning_rate': 3e-05, 'epoch': 7.08}
|
| 371 |
-
{'loss': 0.3096, 'grad_norm': 0.6403458118438721, 'learning_rate': 2.9714285714285717e-05, 'epoch': 7.11}
|
| 372 |
-
{'loss': 0.3103, 'grad_norm': 0.7583682537078857, 'learning_rate': 2.9428571428571426e-05, 'epoch': 7.14}
|
| 373 |
-
{'loss': 0.3064, 'grad_norm': 0.8137710094451904, 'learning_rate': 2.9142857142857146e-05, 'epoch': 7.17}
|
| 374 |
-
{'loss': 0.3067, 'grad_norm': 0.7179896235466003, 'learning_rate': 2.885714285714286e-05, 'epoch': 7.19}
|
| 375 |
-
{'loss': 0.3081, 'grad_norm': 0.9344987273216248, 'learning_rate': 2.857142857142857e-05, 'epoch': 7.22}
|
| 376 |
-
75%|███████████████████████████████████████████████████████████████████████████ | 270/360 [03:28<01:10, 1.28it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 377 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 378 |
-
{'eval_loss': 0.33135512471199036, 'eval_runtime': 0.1054, 'eval_samples_per_second': 436.494, 'eval_steps_per_second': 9.489, 'epoch': 7.22}
|
| 379 |
-
{'loss': 0.3024, 'grad_norm': 0.7846106886863708, 'learning_rate': 2.8285714285714287e-05, 'epoch': 7.25}
|
| 380 |
-
{'loss': 0.3062, 'grad_norm': 0.5884716510772705, 'learning_rate': 2.8000000000000003e-05, 'epoch': 7.28}
|
| 381 |
-
{'loss': 0.3017, 'grad_norm': 0.7277001738548279, 'learning_rate': 2.7714285714285716e-05, 'epoch': 7.31}
|
| 382 |
-
{'loss': 0.3109, 'grad_norm': 0.6671104431152344, 'learning_rate': 2.742857142857143e-05, 'epoch': 7.33}
|
| 383 |
-
{'loss': 0.3051, 'grad_norm': 0.6468051671981812, 'learning_rate': 2.714285714285714e-05, 'epoch': 7.36}
|
| 384 |
-
{'loss': 0.3059, 'grad_norm': 0.7413132190704346, 'learning_rate': 2.6857142857142857e-05, 'epoch': 7.39}
|
| 385 |
-
{'loss': 0.3108, 'grad_norm': 0.8842555284500122, 'learning_rate': 2.6571428571428576e-05, 'epoch': 7.42}
|
| 386 |
-
{'loss': 0.31, 'grad_norm': 0.7701683044433594, 'learning_rate': 2.6285714285714286e-05, 'epoch': 7.44}
|
| 387 |
-
{'loss': 0.2966, 'grad_norm': 0.6261523962020874, 'learning_rate': 2.6000000000000002e-05, 'epoch': 7.47}
|
| 388 |
-
{'loss': 0.3063, 'grad_norm': 0.6180337071418762, 'learning_rate': 2.5714285714285714e-05, 'epoch': 7.5}
|
| 389 |
-
78%|█████████████████████████████████████████████████████████████████████████████▊ | 280/360 [03:36<01:00, 1.32it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 390 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 391 |
-
{'eval_loss': 0.33175382018089294, 'eval_runtime': 0.1062, 'eval_samples_per_second': 433.162, 'eval_steps_per_second': 9.417, 'epoch': 7.5}
|
| 392 |
-
{'loss': 0.3058, 'grad_norm': 0.910743236541748, 'learning_rate': 2.542857142857143e-05, 'epoch': 7.53}
|
| 393 |
-
{'loss': 0.3092, 'grad_norm': 0.8947623372077942, 'learning_rate': 2.5142857142857147e-05, 'epoch': 7.56}
|
| 394 |
-
{'loss': 0.3069, 'grad_norm': 0.8466401100158691, 'learning_rate': 2.485714285714286e-05, 'epoch': 7.58}
|
| 395 |
-
{'loss': 0.316, 'grad_norm': 0.7808226943016052, 'learning_rate': 2.4571428571428572e-05, 'epoch': 7.61}
|
| 396 |
-
{'loss': 0.2993, 'grad_norm': 0.6704230308532715, 'learning_rate': 2.4285714285714288e-05, 'epoch': 7.64}
|
| 397 |
-
{'loss': 0.3034, 'grad_norm': 0.7090719938278198, 'learning_rate': 2.4e-05, 'epoch': 7.67}
|
| 398 |
-
{'loss': 0.3077, 'grad_norm': 0.7552341818809509, 'learning_rate': 2.3714285714285717e-05, 'epoch': 7.69}
|
| 399 |
-
{'loss': 0.3047, 'grad_norm': 0.7747870683670044, 'learning_rate': 2.342857142857143e-05, 'epoch': 7.72}
|
| 400 |
-
{'loss': 0.3105, 'grad_norm': 1.1127567291259766, 'learning_rate': 2.3142857142857145e-05, 'epoch': 7.75}
|
| 401 |
-
{'loss': 0.2997, 'grad_norm': 0.7083399891853333, 'learning_rate': 2.2857142857142858e-05, 'epoch': 7.78}
|
| 402 |
-
81%|████████████████████████████████████████████████████████████████████████████████▌ | 290/360 [03:43<00:50, 1.40it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 403 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 404 |
-
{'eval_loss': 0.3296756446361542, 'eval_runtime': 0.1057, 'eval_samples_per_second': 435.219, 'eval_steps_per_second': 9.461, 'epoch': 7.78}
|
| 405 |
-
{'loss': 0.3035, 'grad_norm': 0.5560160279273987, 'learning_rate': 2.257142857142857e-05, 'epoch': 7.81}
|
| 406 |
-
{'loss': 0.3066, 'grad_norm': 0.7277750372886658, 'learning_rate': 2.2285714285714287e-05, 'epoch': 7.83}
|
| 407 |
-
{'loss': 0.3049, 'grad_norm': 0.683189868927002, 'learning_rate': 2.2000000000000003e-05, 'epoch': 7.86}
|
| 408 |
-
{'loss': 0.3098, 'grad_norm': 0.7173867225646973, 'learning_rate': 2.1714285714285715e-05, 'epoch': 7.89}
|
| 409 |
-
{'loss': 0.3081, 'grad_norm': 0.698379397392273, 'learning_rate': 2.1428571428571428e-05, 'epoch': 7.92}
|
| 410 |
-
{'loss': 0.3109, 'grad_norm': 0.5789396166801453, 'learning_rate': 2.1142857142857144e-05, 'epoch': 7.94}
|
| 411 |
-
{'loss': 0.3149, 'grad_norm': 0.8007158637046814, 'learning_rate': 2.0857142857142857e-05, 'epoch': 7.97}
|
| 412 |
-
{'loss': 0.3087, 'grad_norm': 0.8773916959762573, 'learning_rate': 2.0571428571428573e-05, 'epoch': 8.0}
|
| 413 |
-
{'loss': 0.2989, 'grad_norm': 0.7212807536125183, 'learning_rate': 2.0285714285714286e-05, 'epoch': 8.03}
|
| 414 |
-
{'loss': 0.3069, 'grad_norm': 0.6096076369285583, 'learning_rate': 2e-05, 'epoch': 8.06}
|
| 415 |
-
83%|███████████████████████████████████████████████████████████████████████████████████▎ | 300/360 [03:51<00:46, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 416 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 417 |
-
{'eval_loss': 0.33069342374801636, 'eval_runtime': 0.105, 'eval_samples_per_second': 438.062, 'eval_steps_per_second': 9.523, 'epoch': 8.06}
|
| 418 |
-
{'loss': 0.295, 'grad_norm': 0.5173125267028809, 'learning_rate': 1.9714285714285714e-05, 'epoch': 8.08}
|
| 419 |
-
{'loss': 0.3014, 'grad_norm': 0.6913369297981262, 'learning_rate': 1.942857142857143e-05, 'epoch': 8.11}
|
| 420 |
-
{'loss': 0.3037, 'grad_norm': 0.7195921540260315, 'learning_rate': 1.9142857142857143e-05, 'epoch': 8.14}
|
| 421 |
-
{'loss': 0.3031, 'grad_norm': 0.6366473436355591, 'learning_rate': 1.885714285714286e-05, 'epoch': 8.17}
|
| 422 |
-
{'loss': 0.2957, 'grad_norm': 0.5457173585891724, 'learning_rate': 1.8571428571428572e-05, 'epoch': 8.19}
|
| 423 |
-
{'loss': 0.2997, 'grad_norm': 0.6149912476539612, 'learning_rate': 1.8285714285714288e-05, 'epoch': 8.22}
|
| 424 |
-
{'loss': 0.3048, 'grad_norm': 0.5352884531021118, 'learning_rate': 1.8e-05, 'epoch': 8.25}
|
| 425 |
-
{'loss': 0.308, 'grad_norm': 0.6278409361839294, 'learning_rate': 1.7714285714285713e-05, 'epoch': 8.28}
|
| 426 |
-
{'loss': 0.3005, 'grad_norm': 0.5881698727607727, 'learning_rate': 1.742857142857143e-05, 'epoch': 8.31}
|
| 427 |
-
{'loss': 0.302, 'grad_norm': 0.6125136613845825, 'learning_rate': 1.7142857142857145e-05, 'epoch': 8.33}
|
| 428 |
-
86%|██████████████████████████████████████████████████████████████████████████████████████ | 310/360 [03:59<00:38, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 429 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 430 |
-
{'eval_loss': 0.3336547017097473, 'eval_runtime': 0.1055, 'eval_samples_per_second': 436.2, 'eval_steps_per_second': 9.483, 'epoch': 8.33}
|
| 431 |
-
{'loss': 0.3012, 'grad_norm': 0.6722866892814636, 'learning_rate': 1.6857142857142858e-05, 'epoch': 8.36}
|
| 432 |
-
{'loss': 0.297, 'grad_norm': 0.6827422976493835, 'learning_rate': 1.657142857142857e-05, 'epoch': 8.39}
|
| 433 |
-
{'loss': 0.2977, 'grad_norm': 0.7612675428390503, 'learning_rate': 1.6285714285714287e-05, 'epoch': 8.42}
|
| 434 |
-
{'loss': 0.3009, 'grad_norm': 0.5952971577644348, 'learning_rate': 1.6000000000000003e-05, 'epoch': 8.44}
|
| 435 |
-
{'loss': 0.3043, 'grad_norm': 0.8323265314102173, 'learning_rate': 1.5714285714285715e-05, 'epoch': 8.47}
|
| 436 |
-
{'loss': 0.2956, 'grad_norm': 0.8321357369422913, 'learning_rate': 1.5428571428571428e-05, 'epoch': 8.5}
|
| 437 |
-
{'loss': 0.3029, 'grad_norm': 0.6457182168960571, 'learning_rate': 1.5142857142857144e-05, 'epoch': 8.53}
|
| 438 |
-
{'loss': 0.2972, 'grad_norm': 0.5753086805343628, 'learning_rate': 1.4857142857142858e-05, 'epoch': 8.56}
|
| 439 |
-
{'loss': 0.2966, 'grad_norm': 0.8767444491386414, 'learning_rate': 1.4571428571428573e-05, 'epoch': 8.58}
|
| 440 |
-
{'loss': 0.3049, 'grad_norm': 0.929669201374054, 'learning_rate': 1.4285714285714285e-05, 'epoch': 8.61}
|
| 441 |
-
89%|████████████████████████████████████████████████████████████████████████████████████████▉ | 320/360 [04:07<00:31, 1.29it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 442 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 443 |
-
{'eval_loss': 0.3303754925727844, 'eval_runtime': 0.1057, 'eval_samples_per_second': 435.106, 'eval_steps_per_second': 9.459, 'epoch': 8.61}
|
| 444 |
-
{'loss': 0.2989, 'grad_norm': 0.7576697468757629, 'learning_rate': 1.4000000000000001e-05, 'epoch': 8.64}
|
| 445 |
-
{'loss': 0.3051, 'grad_norm': 0.6402246952056885, 'learning_rate': 1.3714285714285716e-05, 'epoch': 8.67}
|
| 446 |
-
{'loss': 0.2974, 'grad_norm': 0.5665248036384583, 'learning_rate': 1.3428571428571429e-05, 'epoch': 8.69}
|
| 447 |
-
{'loss': 0.3061, 'grad_norm': 0.9747456312179565, 'learning_rate': 1.3142857142857143e-05, 'epoch': 8.72}
|
| 448 |
-
{'loss': 0.3012, 'grad_norm': 0.657123863697052, 'learning_rate': 1.2857142857142857e-05, 'epoch': 8.75}
|
| 449 |
-
{'loss': 0.2995, 'grad_norm': 0.7186892032623291, 'learning_rate': 1.2571428571428573e-05, 'epoch': 8.78}
|
| 450 |
-
{'loss': 0.3026, 'grad_norm': 0.6889364123344421, 'learning_rate': 1.2285714285714286e-05, 'epoch': 8.81}
|
| 451 |
-
{'loss': 0.3009, 'grad_norm': 0.6299145817756653, 'learning_rate': 1.2e-05, 'epoch': 8.83}
|
| 452 |
-
{'loss': 0.3002, 'grad_norm': 0.7328559756278992, 'learning_rate': 1.1714285714285715e-05, 'epoch': 8.86}
|
| 453 |
-
{'loss': 0.2953, 'grad_norm': 0.6111913919448853, 'learning_rate': 1.1428571428571429e-05, 'epoch': 8.89}
|
| 454 |
-
92%|███████████████████████████████████████████████████████████████████████████████████████████▋ | 330/360 [04:14<00:22, 1.31it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 455 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 456 |
-
{'eval_loss': 0.3253832757472992, 'eval_runtime': 0.106, 'eval_samples_per_second': 434.082, 'eval_steps_per_second': 9.437, 'epoch': 8.89}
|
| 457 |
-
{'loss': 0.3023, 'grad_norm': 0.6739629507064819, 'learning_rate': 1.1142857142857143e-05, 'epoch': 8.92}
|
| 458 |
-
{'loss': 0.2978, 'grad_norm': 0.6967675685882568, 'learning_rate': 1.0857142857142858e-05, 'epoch': 8.94}
|
| 459 |
-
{'loss': 0.3043, 'grad_norm': 0.702989935874939, 'learning_rate': 1.0571428571428572e-05, 'epoch': 8.97}
|
| 460 |
-
{'loss': 0.3034, 'grad_norm': 1.156525731086731, 'learning_rate': 1.0285714285714286e-05, 'epoch': 9.0}
|
| 461 |
-
{'loss': 0.2989, 'grad_norm': 0.564460277557373, 'learning_rate': 1e-05, 'epoch': 9.03}
|
| 462 |
-
{'loss': 0.2955, 'grad_norm': 0.5435044169425964, 'learning_rate': 9.714285714285715e-06, 'epoch': 9.06}
|
| 463 |
-
{'loss': 0.2986, 'grad_norm': 0.511762797832489, 'learning_rate': 9.42857142857143e-06, 'epoch': 9.08}
|
| 464 |
-
{'loss': 0.2932, 'grad_norm': 0.6208844780921936, 'learning_rate': 9.142857142857144e-06, 'epoch': 9.11}
|
| 465 |
-
{'loss': 0.2932, 'grad_norm': 0.5209355354309082, 'learning_rate': 8.857142857142857e-06, 'epoch': 9.14}
|
| 466 |
-
{'loss': 0.302, 'grad_norm': 0.5852081775665283, 'learning_rate': 8.571428571428573e-06, 'epoch': 9.17}
|
| 467 |
-
94%|██████████████████████████████████████████████████████████████████████████████████████████████▍ | 340/360 [04:22<00:15, 1.28it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 468 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 469 |
-
{'eval_loss': 0.32703322172164917, 'eval_runtime': 0.1055, 'eval_samples_per_second': 435.944, 'eval_steps_per_second': 9.477, 'epoch': 9.17}
|
| 470 |
-
{'loss': 0.2973, 'grad_norm': 0.6613155603408813, 'learning_rate': 8.285714285714285e-06, 'epoch': 9.19}
|
| 471 |
-
{'loss': 0.296, 'grad_norm': 0.6458805203437805, 'learning_rate': 8.000000000000001e-06, 'epoch': 9.22}
|
| 472 |
-
{'loss': 0.2998, 'grad_norm': 0.5602886080741882, 'learning_rate': 7.714285714285714e-06, 'epoch': 9.25}
|
| 473 |
-
{'loss': 0.2898, 'grad_norm': 0.5723817348480225, 'learning_rate': 7.428571428571429e-06, 'epoch': 9.28}
|
| 474 |
-
{'loss': 0.2935, 'grad_norm': 0.6257355213165283, 'learning_rate': 7.142857142857143e-06, 'epoch': 9.31}
|
| 475 |
-
{'loss': 0.2909, 'grad_norm': 0.6624913811683655, 'learning_rate': 6.857142857142858e-06, 'epoch': 9.33}
|
| 476 |
-
{'loss': 0.3001, 'grad_norm': 0.5716632604598999, 'learning_rate': 6.5714285714285714e-06, 'epoch': 9.36}
|
| 477 |
-
{'loss': 0.2964, 'grad_norm': 0.6996496319770813, 'learning_rate': 6.285714285714287e-06, 'epoch': 9.39}
|
| 478 |
-
{'loss': 0.2927, 'grad_norm': 0.7235862612724304, 'learning_rate': 6e-06, 'epoch': 9.42}
|
| 479 |
-
{'loss': 0.2956, 'grad_norm': 0.6455687284469604, 'learning_rate': 5.7142857142857145e-06, 'epoch': 9.44}
|
| 480 |
-
97%|█████████████████████████████████████████████████████████████████████████████████████████████████▏ | 350/360 [04:30<00:07, 1.28it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 481 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 482 |
-
{'eval_loss': 0.32588478922843933, 'eval_runtime': 0.1057, 'eval_samples_per_second': 435.067, 'eval_steps_per_second': 9.458, 'epoch': 9.44}
|
| 483 |
-
{'loss': 0.2959, 'grad_norm': 0.5984405279159546, 'learning_rate': 5.428571428571429e-06, 'epoch': 9.47}
|
| 484 |
-
{'loss': 0.2936, 'grad_norm': 0.6240008473396301, 'learning_rate': 5.142857142857143e-06, 'epoch': 9.5}
|
| 485 |
-
{'loss': 0.2987, 'grad_norm': 0.6871291399002075, 'learning_rate': 4.857142857142858e-06, 'epoch': 9.53}
|
| 486 |
-
{'loss': 0.2941, 'grad_norm': 0.6369628310203552, 'learning_rate': 4.571428571428572e-06, 'epoch': 9.56}
|
| 487 |
-
{'loss': 0.2959, 'grad_norm': 0.6651211977005005, 'learning_rate': 4.285714285714286e-06, 'epoch': 9.58}
|
| 488 |
-
{'loss': 0.2991, 'grad_norm': 0.7005758285522461, 'learning_rate': 4.000000000000001e-06, 'epoch': 9.61}
|
| 489 |
-
{'loss': 0.2936, 'grad_norm': 0.5685088634490967, 'learning_rate': 3.7142857142857146e-06, 'epoch': 9.64}
|
| 490 |
-
{'loss': 0.2947, 'grad_norm': 0.6322896480560303, 'learning_rate': 3.428571428571429e-06, 'epoch': 9.67}
|
| 491 |
-
{'loss': 0.2951, 'grad_norm': 0.6149244904518127, 'learning_rate': 3.1428571428571433e-06, 'epoch': 9.69}
|
| 492 |
-
{'loss': 0.2975, 'grad_norm': 0.685043215751648, 'learning_rate': 2.8571428571428573e-06, 'epoch': 9.72}
|
| 493 |
-
100%|████████████████████████████████████████████████████████████████████████████████████████████████████| 360/360 [04:37<00:00, 1.54it/s]Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 494 |
-
Trainer.tokenizer is now deprecated. You should use Trainer.processing_class instead.
|
| 495 |
-
{'eval_loss': 0.3277616798877716, 'eval_runtime': 0.1059, 'eval_samples_per_second': 434.21, 'eval_steps_per_second': 9.439, 'epoch': 9.72}
|
| 496 |
-
{'loss': 0.2951, 'grad_norm': 0.8039355874061584, 'learning_rate': 2.5714285714285716e-06, 'epoch': 9.75}
|
| 497 |
-
{'loss': 0.2928, 'grad_norm': 0.6388571262359619, 'learning_rate': 2.285714285714286e-06, 'epoch': 9.78}
|
| 498 |
-
{'loss': 0.2934, 'grad_norm': 0.5934285521507263, 'learning_rate': 2.0000000000000003e-06, 'epoch': 9.81}
|
| 499 |
-
{'loss': 0.2952, 'grad_norm': 0.5320731401443481, 'learning_rate': 1.7142857142857145e-06, 'epoch': 9.83}
|
| 500 |
-
{'loss': 0.2932, 'grad_norm': 0.6137614846229553, 'learning_rate': 1.4285714285714286e-06, 'epoch': 9.86}
|
| 501 |
-
{'loss': 0.2944, 'grad_norm': 0.8172494769096375, 'learning_rate': 1.142857142857143e-06, 'epoch': 9.89}
|
| 502 |
-
{'loss': 0.2917, 'grad_norm': 0.6931514739990234, 'learning_rate': 8.571428571428572e-07, 'epoch': 9.92}
|
| 503 |
-
{'loss': 0.2952, 'grad_norm': 0.8408763408660889, 'learning_rate': 5.714285714285715e-07, 'epoch': 9.94}
|
| 504 |
-
{'loss': 0.2948, 'grad_norm': 0.6687312126159668, 'learning_rate': 2.8571428571428575e-07, 'epoch': 9.97}
|
| 505 |
-
{'loss': 0.2944, 'grad_norm': 0.8545575737953186, 'learning_rate': 0.0, 'epoch': 10.0}
|
| 506 |
-
100%|████████████████████████████████████████████████████████████████████████████████████████████████████| 360/360 [04:39<00:00, 1.29it/s]
|
| 507 |
-
|
| 508 |
-
{'eval_loss': 0.3285914361476898, 'eval_runtime': 0.1058, 'eval_samples_per_second': 434.729, 'eval_steps_per_second': 9.451, 'epoch': 10.0}
|
| 509 |
-
{'train_runtime': 279.4567, 'train_samples_per_second': 162.422, 'train_steps_per_second': 1.288, 'train_loss': 0.35597237489289707, 'epoch': 10.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/files/wandb-metadata.json
DELETED
|
@@ -1,55 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"os": "Linux-6.8.0-101-generic-x86_64-with-glibc2.35",
|
| 3 |
-
"python": "3.9.20",
|
| 4 |
-
"startedAt": "2026-03-27T08:20:56.140561Z",
|
| 5 |
-
"args": [
|
| 6 |
-
"--config-path",
|
| 7 |
-
"config/main_table",
|
| 8 |
-
"--config-name",
|
| 9 |
-
"llmdp_llm_adroit-hand-hammer-v1.yaml"
|
| 10 |
-
],
|
| 11 |
-
"program": "/tmp2/chyang/workspace/LLM-BC/./train.py",
|
| 12 |
-
"codePath": "train.py",
|
| 13 |
-
"git": {
|
| 14 |
-
"remote": "https://github.com/CHYang25/LLM-BC.git",
|
| 15 |
-
"commit": "2e85824cdb13f64f31923d9430e890dadc78d394"
|
| 16 |
-
},
|
| 17 |
-
"email": "chris920325@gmail.com",
|
| 18 |
-
"root": "/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1",
|
| 19 |
-
"host": "A6000-2",
|
| 20 |
-
"username": "chyang",
|
| 21 |
-
"executable": "/home/chyang/miniconda3/envs/llm-bc/bin/python3",
|
| 22 |
-
"codePathLocal": "train.py",
|
| 23 |
-
"cpu_count": 12,
|
| 24 |
-
"cpu_count_logical": 24,
|
| 25 |
-
"gpu": "NVIDIA RTX A6000",
|
| 26 |
-
"gpu_count": 2,
|
| 27 |
-
"disk": {
|
| 28 |
-
"/": {
|
| 29 |
-
"total": "1967317549056",
|
| 30 |
-
"used": "733081182208"
|
| 31 |
-
}
|
| 32 |
-
},
|
| 33 |
-
"memory": {
|
| 34 |
-
"total": "134538502144"
|
| 35 |
-
},
|
| 36 |
-
"cpu": {
|
| 37 |
-
"count": 12,
|
| 38 |
-
"countLogical": 24
|
| 39 |
-
},
|
| 40 |
-
"gpu_nvidia": [
|
| 41 |
-
{
|
| 42 |
-
"name": "NVIDIA RTX A6000",
|
| 43 |
-
"memoryTotal": "51527024640",
|
| 44 |
-
"cudaCores": 10752,
|
| 45 |
-
"architecture": "Ampere"
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"name": "NVIDIA RTX A6000",
|
| 49 |
-
"memoryTotal": "51527024640",
|
| 50 |
-
"cudaCores": 10752,
|
| 51 |
-
"architecture": "Ampere"
|
| 52 |
-
}
|
| 53 |
-
],
|
| 54 |
-
"cudaVersion": "12.6"
|
| 55 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/files/wandb-summary.json
DELETED
|
@@ -1 +0,0 @@
|
|
| 1 |
-
{"eval/loss":0.3285914361476898,"train_loss":0.35597237489289707,"eval/steps_per_second":9.451,"train/global_step":360,"train/grad_norm":0.8545575737953186,"train/learning_rate":0,"_timestamp":1.774599943812631e+09,"_step":396,"eval/runtime":0.1058,"train_steps_per_second":1.288,"eval/samples_per_second":434.729,"_runtime":287.672274788,"train_samples_per_second":162.422,"train/loss":0.2944,"_wandb":{"runtime":287},"train_runtime":279.4567,"train/epoch":10,"total_flos":3.83135455600128e+15}
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug-core.log
DELETED
|
@@ -1,16 +0,0 @@
|
|
| 1 |
-
{"time":"2026-03-27T16:20:55.459583429+08:00","level":"INFO","msg":"started logging, with flags","port-filename":"/tmp/tmp0ig_2z3y/port-2703097.txt","pid":2703097,"debug":false,"disable-analytics":false}
|
| 2 |
-
{"time":"2026-03-27T16:20:55.4596054+08:00","level":"INFO","msg":"FeatureState","shutdownOnParentExitEnabled":false}
|
| 3 |
-
{"time":"2026-03-27T16:20:55.460720879+08:00","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":44997,"Zone":""}}
|
| 4 |
-
{"time":"2026-03-27T16:20:55.460793184+08:00","level":"INFO","msg":"Will exit if parent process dies.","ppid":2703097}
|
| 5 |
-
{"time":"2026-03-27T16:20:55.657444973+08:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:60248"}
|
| 6 |
-
{"time":"2026-03-27T16:20:56.143936837+08:00","level":"INFO","msg":"handleInformInit: received","streamId":"nhmfpc2t","id":"127.0.0.1:60248"}
|
| 7 |
-
{"time":"2026-03-27T16:20:56.25130017+08:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"nhmfpc2t","id":"127.0.0.1:60248"}
|
| 8 |
-
{"time":"2026-03-27T16:25:50.90098896+08:00","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"nhmfpc2t","id":"127.0.0.1:60248"}
|
| 9 |
-
{"time":"2026-03-27T16:25:50.901237575+08:00","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"nhmfpc2t","id":"127.0.0.1:60248"}
|
| 10 |
-
{"time":"2026-03-27T16:25:51.590522527+08:00","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:60248"}
|
| 11 |
-
{"time":"2026-03-27T16:25:51.59056499+08:00","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:60248"}
|
| 12 |
-
{"time":"2026-03-27T16:25:51.590573489+08:00","level":"INFO","msg":"server is shutting down"}
|
| 13 |
-
{"time":"2026-03-27T16:25:51.590600946+08:00","level":"INFO","msg":"connection: Close: initiating connection closure","id":"127.0.0.1:60248"}
|
| 14 |
-
{"time":"2026-03-27T16:25:51.590777149+08:00","level":"INFO","msg":"connection: Close: connection successfully closed","id":"127.0.0.1:60248"}
|
| 15 |
-
{"time":"2026-03-27T16:25:51.590829802+08:00","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:60248"}
|
| 16 |
-
{"time":"2026-03-27T16:25:51.590853327+08:00","level":"INFO","msg":"server is closed"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug-internal.log
DELETED
|
@@ -1,17 +0,0 @@
|
|
| 1 |
-
{"time":"2026-03-27T16:20:56.144221699+08:00","level":"INFO","msg":"using version","core version":"0.18.6"}
|
| 2 |
-
{"time":"2026-03-27T16:20:56.144248456+08:00","level":"INFO","msg":"created symlink","path":"/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug-core.log"}
|
| 3 |
-
{"time":"2026-03-27T16:20:56.25122394+08:00","level":"INFO","msg":"created new stream","id":"nhmfpc2t"}
|
| 4 |
-
{"time":"2026-03-27T16:20:56.25128922+08:00","level":"INFO","msg":"stream: started","id":"nhmfpc2t"}
|
| 5 |
-
{"time":"2026-03-27T16:20:56.251326647+08:00","level":"INFO","msg":"sender: started","stream_id":"nhmfpc2t"}
|
| 6 |
-
{"time":"2026-03-27T16:20:56.251319685+08:00","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"nhmfpc2t"}}
|
| 7 |
-
{"time":"2026-03-27T16:20:56.251303072+08:00","level":"INFO","msg":"handler: started","stream_id":{"value":"nhmfpc2t"}}
|
| 8 |
-
{"time":"2026-03-27T16:20:57.417356613+08:00","level":"INFO","msg":"Starting system monitor"}
|
| 9 |
-
{"time":"2026-03-27T16:25:43.814415705+08:00","level":"INFO","msg":"Stopping system monitor"}
|
| 10 |
-
{"time":"2026-03-27T16:25:43.815067495+08:00","level":"INFO","msg":"Stopped system monitor"}
|
| 11 |
-
{"time":"2026-03-27T16:25:44.815119947+08:00","level":"INFO","msg":"handler: operation stats","stats":{"operations":[{"desc":"uploading wandb-summary.json","runtime_seconds":0.136938035,"progress":"495B/495B"},{"desc":"saving job artifact","runtime_seconds":0.037243753}],"total_operations":2}}
|
| 12 |
-
{"time":"2026-03-27T16:25:49.240469481+08:00","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
|
| 13 |
-
{"time":"2026-03-27T16:25:50.901076411+08:00","level":"INFO","msg":"stream: closing","id":"nhmfpc2t"}
|
| 14 |
-
{"time":"2026-03-27T16:25:50.901107424+08:00","level":"INFO","msg":"handler: closed","stream_id":{"value":"nhmfpc2t"}}
|
| 15 |
-
{"time":"2026-03-27T16:25:50.901135119+08:00","level":"INFO","msg":"writer: Close: closed","stream_id":{"value":"nhmfpc2t"}}
|
| 16 |
-
{"time":"2026-03-27T16:25:50.901152717+08:00","level":"INFO","msg":"sender: closed","stream_id":"nhmfpc2t"}
|
| 17 |
-
{"time":"2026-03-27T16:25:50.90122351+08:00","level":"INFO","msg":"stream: closed","id":"nhmfpc2t"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug.log
DELETED
|
@@ -1,35 +0,0 @@
|
|
| 1 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Current SDK version is 0.18.6
|
| 2 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Configure stats pid to 2703097
|
| 3 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Loading settings from /home/chyang/.config/wandb/settings
|
| 4 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Loading settings from /tmp2/chyang/workspace/LLM-BC/wandb/settings
|
| 5 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Loading settings from environment variables: {}
|
| 6 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': 'online', '_disable_service': None}
|
| 7 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train.py', 'program_abspath': '/tmp2/chyang/workspace/LLM-BC/train.py', 'program': '/tmp2/chyang/workspace/LLM-BC/./train.py'}
|
| 8 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_setup.py:_flush():79] Applying login settings: {}
|
| 9 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:_log_setup():533] Logging user logs to /tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug.log
|
| 10 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:_log_setup():534] Logging internal logs to /tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/logs/debug-internal.log
|
| 11 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:init():619] calling init triggers
|
| 12 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:init():626] wandb.init called with sweep_config: {}
|
| 13 |
-
config: {'name': 'train_llm_lowdim', '_target_': 'llmbc.workspace.train_llm_workspace.TrainLLMWorkspace', 'obs_dim': 46, 'action_dim': 26, 'horizon': 1, 'n_obs_steps': 1, 'n_action_steps': 1, 'task_name': 'adroit-hand-hammer-v1', 'exp_name': 'train llm', 'model_name': 'HuggingFaceTB/SmolLM2-135M-Instruct', 'use_quantization': False, 'lora_config': {'r': 32, 'lora_alpha': 64, 'lora_dropout': 0.05, 'bias': 'none', 'task_type': 'CAUSAL_LM'}, 'dataset': {'test_data_ratio': 0.01}, 'debug': False, 'training': {'seed': 42, 'per_device_train_batch_size': 128, 'per_device_eval_batch_size': 128, 'gradient_accumulation_steps': 1, 'optim': 'paged_adamw_32bit', 'num_train_epochs': 10, 'eval_strategy': 'steps', 'logging_steps': 1, 'warmup_steps': 10, 'logging_strategy': 'steps', 'learning_rate': 0.0001, 'fp16': False, 'bf16': True, 'tf32': True, 'group_by_length': True, 'report_to': 'wandb', 'save_steps': 5000, 'eval_steps': 10, 'use_joint_mlp_projector': True, 'joint_obs_action_mlp_lr': 5e-05}, 'trainer': {'obs_dim': 46, 'action_dim': 26, 'use_joint_mlp_projector': True, 'max_seq_length': 100, 'dataset_text_field': 'text', 'packing': False}, 'logging': {'project': 'llm_module_finetuning', 'resume': True, 'mode': 'online', 'name': '2026.03.27-16.20.52_train_llm_lowdim_adroit-hand-hammer-v1', 'tags': ['train_llm_lowdim', 'adroit-hand-hammer-v1', 'train llm'], 'id': None, 'group': None}, 'multi_run': {'run_dir': 'data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1', 'wandb_name_base': '2026.03.27-16.20.52_train_llm_lowdim_adroit-hand-hammer-v1'}, 'task': {'name': 'adroit-hand-hammer-v1', 'obs_dim': 46, 'action_dim': 26, 'env_runner': {'_target_': 'llmbc.env_runner.adroit_lowdim_runner.AdroitHandLowdimRunner', 'env_name': 'llf-adroit-adroit-hand-hammer-v1', 'n_train': 10, 'n_test': 50, 'n_envs': 10, 'max_steps': 150, 'n_obs_steps': 1, 'n_action_steps': 1, 'instruction_type': 'b', 'feedback_type': ['hp', 'hn', 'fp'], 'visual': False, 'discount': 0.99}, 'dataset': {'_target_': 'llmbc.dataset.adroit_lowdim_dataset.AdroitHandLowdimDataset', 'data_path': 'datasets/adroit-hand-hammer-v1-general.pt', 'data_path2': 'datasets/adroit-hand-hammer-v1.pt', 'horizon': 1, 'pad_before': 0, 'pad_after': 0, 'obs_eef_target': True, 'use_manual_normalizer': False, 'val_ratio': 0.05, 'dummy_normalizer': False}, 'instructor': {'_target_': 'llmbc.translator.instructor.adroit_instructor.adroit_hand_hammer_v1_instructor.AdroitHandHammerV1Instructor'}}, 'llm': {'name': 'HuggingFaceTB/SmolLM2-135M-Instruct', 'model_name': 'SmolLM2-135M-Instruct', 'config_target': 'llmbc.model.llm.llama_lowdim_model.LowdimLlamaConfig', 'causal_lm_target': 'llmbc.model.llm.llama_lowdim_model.LowdimLlamaForCausalLM', 'use_quantization': False, 'use_joint_mlp_projector': True, 'llm_mode': 'mlp-finetuned', 'finetune_mode': 'orig', 'checkpoint': 'data/outputs/2026.03.27/14.38.20_train_mlp_projector_adroit-hand-hammer-v1/checkpoints/latest.ckpt', 'max_length': 100, 'lora_config': {'r': 32, 'lora_alpha': 64, 'lora_dropout': 0.05, 'bias': 'none', 'task_type': 'CAUSAL_LM'}, 'prompter': {'_target_': 'llmbc.translator.prompter.smollm2_prompter.SmolLM2Prompter', 'use_joint_mlp_projector': True}, 'hydra': {'job': {'override_dirname': 'HuggingFaceTB/SmolLM2-135M-Instruct'}, 'run': {'dir': 'data/outputs/2026.03.27/16.20.52_HuggingFaceTB/SmolLM2-135M-Instruct'}}}}
|
| 14 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:init():669] starting backend
|
| 15 |
-
2026-03-27 16:20:56,139 INFO MainThread:2703097 [wandb_init.py:init():673] sending inform_init request
|
| 16 |
-
2026-03-27 16:20:56,140 INFO MainThread:2703097 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
|
| 17 |
-
2026-03-27 16:20:56,140 INFO MainThread:2703097 [wandb_init.py:init():686] backend started and connected
|
| 18 |
-
2026-03-27 16:20:56,144 INFO MainThread:2703097 [wandb_init.py:init():781] updated telemetry
|
| 19 |
-
2026-03-27 16:20:56,170 INFO MainThread:2703097 [wandb_init.py:init():814] communicating run to backend with 90.0 second timeout
|
| 20 |
-
2026-03-27 16:20:57,414 INFO MainThread:2703097 [wandb_init.py:init():867] starting run threads in backend
|
| 21 |
-
2026-03-27 16:20:57,521 INFO MainThread:2703097 [wandb_run.py:_console_start():2451] atexit reg
|
| 22 |
-
2026-03-27 16:20:57,522 INFO MainThread:2703097 [wandb_run.py:_redirect():2299] redirect: wrap_raw
|
| 23 |
-
2026-03-27 16:20:57,522 INFO MainThread:2703097 [wandb_run.py:_redirect():2364] Wrapping output streams.
|
| 24 |
-
2026-03-27 16:20:57,522 INFO MainThread:2703097 [wandb_run.py:_redirect():2389] Redirects installed.
|
| 25 |
-
2026-03-27 16:20:57,524 INFO MainThread:2703097 [wandb_init.py:init():911] run started, returning control to user process
|
| 26 |
-
2026-03-27 16:21:04,359 INFO MainThread:2703097 [wandb_run.py:_config_callback():1389] config_cb None None {'obs_dim': 46, 'action_dim': 26, 'use_joint_mlp_projector': True, 'vocab_size': 49152, 'max_position_embeddings': 8192, 'hidden_size': 576, 'intermediate_size': 1536, 'num_hidden_layers': 30, 'num_attention_heads': 9, 'num_key_value_heads': 3, 'hidden_act': 'silu', 'initializer_range': 0.041666666666666664, 'rms_norm_eps': 1e-05, 'pretraining_tp': 1, 'use_cache': False, 'rope_theta': 100000, 'rope_scaling': None, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 64, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'bfloat16', 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': None, 'architectures': ['LlamaForCausalLM'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 2, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'HuggingFaceTB/SmolLM2-135M-Instruct', '_attn_implementation_autoset': True, 'transformers_version': '4.47.1', 'is_llama_config': True, 'model_type': 'llama_lowdim', 'rope_interleaved': False, 'transformers.js_config': {'kv_cache_dtype': {'q4f16': 'float16', 'fp16': 'float16'}}, 'output_dir': '/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1', 'overwrite_output_dir': False, 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 128, 'per_device_eval_batch_size': 128, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 0.0001, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 10, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.0, 'warmup_steps': 10, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': '/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1/runs/Mar27_16-21-01_A6000-2', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 5000, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': True, 'fp16': False, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': True, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 10, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': '/tmp2/chyang/workspace/LLM-BC/data/outputs/2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/HuggingFaceTB/SmolLM2-135M-Instruct-finetuned-adroit-hand-hammer-v1', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'paged_adamw_32bit', 'optim_args': None, 'adafactor': False, 'group_by_length': True, 'length_column_name': 'length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'evaluation_strategy': None, 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'dispatch_batches': None, 'split_batches': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'eval_use_gather_object': False, 'average_tokens_across_devices': False, 'dataset_text_field': 'text', 'packing': False, 'max_seq_length': 100, 'dataset_num_proc': None, 'dataset_batch_size': 1000, 'model_init_kwargs': None, 'dataset_kwargs': {}, 'eval_packing': None, 'num_of_sequences': 1024, 'chars_per_token': '<CHARS_PER_TOKEN>', 'use_liger': False, 'joint_obs_action_mlp_lr': 5e-05, 'obs_mlp_lr': None, 'action_mlp_lr': None}
|
| 27 |
-
2026-03-27 16:21:04,361 INFO MainThread:2703097 [wandb_config.py:__setitem__():154] config set model/num_parameters = 134889408 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7e6758745670>>
|
| 28 |
-
2026-03-27 16:21:04,361 INFO MainThread:2703097 [wandb_run.py:_config_callback():1389] config_cb model/num_parameters 134889408 None
|
| 29 |
-
2026-03-27 16:25:43,813 INFO MainThread:2703097 [wandb_run.py:_finish():2146] finishing run chyang25-national-taiwan-university/llm_module_finetuning/nhmfpc2t
|
| 30 |
-
2026-03-27 16:25:43,813 INFO MainThread:2703097 [wandb_run.py:_atexit_cleanup():2414] got exitcode: 0
|
| 31 |
-
2026-03-27 16:25:43,813 INFO MainThread:2703097 [wandb_run.py:_restore():2396] restore
|
| 32 |
-
2026-03-27 16:25:43,814 INFO MainThread:2703097 [wandb_run.py:_restore():2402] restore done
|
| 33 |
-
2026-03-27 16:25:50,896 INFO MainThread:2703097 [wandb_run.py:_footer_history_summary_info():3963] rendering history
|
| 34 |
-
2026-03-27 16:25:50,896 INFO MainThread:2703097 [wandb_run.py:_footer_history_summary_info():3995] rendering summary
|
| 35 |
-
2026-03-27 16:25:50,900 INFO MainThread:2703097 [wandb_run.py:_footer_sync_info():3922] logging synced files
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026.03.27/16.20.52_train_llm_lowdim_adroit-hand-hammer-v1/wandb/run-20260327_162056-nhmfpc2t/run-nhmfpc2t.wandb
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:19281f2406b878cd01e76af72f5dd82f13698026367ccc8dd2b76d39269c8f31
|
| 3 |
-
size 858565
|
|
|
|
|
|
|
|
|
|
|
|