Hugging Face's logo

OpenReasonAI
/
Graphite1.0-4B

Text Generation
PEFT
Safetensors
English
Russian
qwen
qwen3.5
lora
unsloth
trl
sft
code
reasoning
bilingual
obsidian
graphite
conversational
Model card Files
xet
 Community
Starred commited on
Commit
e8ac637
·
verified ·
1 Parent(s): 6687ef8

Training in progress, step 2256, checkpoint

Browse files
Files changed (4) hide show
  1. last-checkpoint/adapter_model.safetensors +1 -1
  2. last-checkpoint/optimizer.pt +1 -1
  3. last-checkpoint/scheduler.pt +1 -1
  4. last-checkpoint/trainer_state.json +4 -4
last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:afc9e576ce294178c6b2fd6e738300f820db1a7fac04d47b3eed7d1fe7b964da
3
  size 84972248
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56a3911c8195f07e7729206b22c78ea087b1f7bcd68b544aea4342793450bd94
3
  size 84972248
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2ad1ba58e3875c031c561de720dee24bbcaf2368184170ad0d5e17db4ea54f8e
3
  size 43434405
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62d491b9b99145a34586738989b18f1697bce28374cbfc95c2b44ca1eecbf851
3
  size 43434405
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8c231d2fef38a21f0b802526519cca758a148fd1c0a091bc4593caa9fef72be0
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83a70477d2c42e292df2c1adf148bed9afdcc2369cdfb05931fd61094ec7740f
3
  size 1465
last-checkpoint/trainer_state.json CHANGED
@@ -2,9 +2,9 @@
2
  "best_global_step": 2250,
3
  "best_metric": 0.18876151740550995,
4
  "best_model_checkpoint": "/kaggle/working/obsidian_critic_qwen35_t4x2_unsloth/runs/obsidian_critic_full_epoch/checkpoint-2250",
5
- "epoch": 0.9977273987029543,
6
  "eval_steps": 125,
7
- "global_step": 2250,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -817,12 +817,12 @@
817
  "should_evaluate": false,
818
  "should_log": false,
819
  "should_save": true,
820
- "should_training_stop": false
821
  },
822
  "attributes": {}
823
  }
824
  },
825
- "total_flos": 1.6621215424289178e+17,
826
  "train_batch_size": 1,
827
  "trial_name": null,
828
  "trial_params": null
 
2
  "best_global_step": 2250,
3
  "best_metric": 0.18876151740550995,
4
  "best_model_checkpoint": "/kaggle/working/obsidian_critic_qwen35_t4x2_unsloth/runs/obsidian_critic_full_epoch/checkpoint-2250",
5
+ "epoch": 1.0,
6
  "eval_steps": 125,
7
+ "global_step": 2256,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
817
  "should_evaluate": false,
818
  "should_log": false,
819
  "should_save": true,
820
+ "should_training_stop": true
821
  },
822
  "attributes": {}
823
  }
824
  },
825
+ "total_flos": 1.666058653049815e+17,
826
  "train_batch_size": 1,
827
  "trial_name": null,
828
  "trial_params": null