Hugging Face's logo

LLM-OS-Models
/
KoHRM-Text-1.4B-raw-checkpoints

Text Generation
Transformers
English
Korean
terminal
sft
vllm
tb2-lite
evaluation-pending
Model card Files
xet
 Community
gyung commited on
Commit
6ed6327
·
verified ·
1 Parent(s): 5ad486b

Add files using upload-large-folder tool

Browse files
Files changed (23) hide show
  1. .gitattributes +9 -0
  2. stage4-korean-tool-finance-step230000/all_config.yaml +46 -0
  3. stage4-korean-tool-finance-step230000/carry_step_230000.0.pt +3 -0
  4. stage4-korean-tool-finance-step230000/carry_step_230000.1.pt +3 -0
  5. stage4-korean-tool-finance-step230000/carry_step_230000.2.pt +3 -0
  6. stage4-korean-tool-finance-step230000/carry_step_230000.3.pt +3 -0
  7. stage4-korean-tool-finance-step230000/carry_step_230000.4.pt +3 -0
  8. stage4-korean-tool-finance-step230000/carry_step_230000.5.pt +3 -0
  9. stage4-korean-tool-finance-step230000/carry_step_230000.6.pt +3 -0
  10. stage4-korean-tool-finance-step230000/carry_step_230000.7.pt +3 -0
  11. stage4-korean-tool-finance-step230000/fsdp2_step_230000/.metadata +3 -0
  12. stage4-korean-tool-finance-step230000/fsdp2_step_230000/__0_0.distcp +3 -0
  13. stage4-korean-tool-finance-step230000/fsdp2_step_230000/__1_0.distcp +3 -0
  14. stage4-korean-tool-finance-step230000/fsdp2_step_230000/__2_0.distcp +3 -0
  15. stage4-korean-tool-finance-step230000/fsdp2_step_230000/__3_0.distcp +3 -0
  16. stage4-korean-tool-finance-step230000/fsdp2_step_230000/__4_0.distcp +3 -0
  17. stage4-korean-tool-finance-step230000/fsdp2_step_230000/__5_0.distcp +3 -0
  18. stage4-korean-tool-finance-step230000/fsdp2_step_230000/__6_0.distcp +3 -0
  19. stage4-korean-tool-finance-step230000/fsdp2_step_230000/__7_0.distcp +3 -0
  20. stage4-korean-tool-finance-step230000/latest_checkpoint.txt +1 -0
  21. stage4-korean-tool-finance-step230000/step_230000_info.json +8 -0
  22. stage4-korean-tool-finance-step230000/train_metadata.yaml +13 -0
  23. stage4-korean-tool-finance-step230000/upload_manifest.json +1 -0
.gitattributes CHANGED
@@ -294,3 +294,12 @@ stage3-local-terminal/fsdp2_epoch_1/__5_0.distcp filter=lfs diff=lfs merge=lfs -
294
  stage3-local-terminal/fsdp2_epoch_1/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
295
  stage3-local-terminal/fsdp2_epoch_1/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
296
  stage3-local-terminal/fsdp2_epoch_1/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
294
  stage3-local-terminal/fsdp2_epoch_1/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
295
  stage3-local-terminal/fsdp2_epoch_1/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
296
  stage3-local-terminal/fsdp2_epoch_1/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
297
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
298
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/.metadata filter=lfs diff=lfs merge=lfs -text
299
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
300
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
301
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
302
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
303
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
304
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
305
+ stage4-korean-tool-finance-step230000/fsdp2_step_230000/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
stage4-korean-tool-finance-step230000/all_config.yaml ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ arch:
2
+ H_cycles: 2
3
+ H_override: {}
4
+ L_cycles: 3
5
+ bp_max_steps: 5
6
+ bp_warmup_ratio: 0.2
7
+ expansion: 4
8
+ half_layers: true
9
+ head: lm_head@LMHead
10
+ hidden_size: 1536
11
+ init_type: lecun_normal
12
+ n_layers: 32
13
+ name: baselines.hrm_nocarry_bp_warmup@HierarchicalReasoningModel
14
+ norm_eps: 1.0e-06
15
+ norm_type: pre
16
+ num_heads: 12
17
+ pos_emb_type: rope
18
+ rope_theta: 10000.0
19
+ beta1: 0.9
20
+ beta2: 0.95
21
+ checkpoint_interval: 1
22
+ checkpoint_keep_last: 2
23
+ checkpoint_path: /home/work/.data/hrm_text_checkpoints/KoHRM-Text-1.4B-stage4-korean-tool-finance-gbs180
24
+ checkpoint_step_interval: 10000
25
+ data:
26
+ path: /home/work/.data/hrm_text_prepared/koterm_korean_tool_finance_mix_v1
27
+ target_only: true
28
+ ema: 0.9999
29
+ epochs: 1
30
+ fwd_bwd_dtype: bfloat16
31
+ global_batch_size: 180224
32
+ log_interval: 5
33
+ lr: 0.00022
34
+ lr_min_ratio: 1.0
35
+ lr_warmup_steps: 2000
36
+ project_name: KoHRM-Text
37
+ resume_epoch: null
38
+ resume_from: /home/work/.data/hrm_text_checkpoints/KoHRM-Text-1.4B-stage3-local-terminal-gbs180
39
+ resume_step: null
40
+ resume_step_offset: 220433
41
+ run_name: KoHRM-Text-1.4B-stage4-korean-tool-finance
42
+ seed: 0
43
+ skip_batches: 0
44
+ total_steps_override: 465000
45
+ weight_decay: 0.1
46
+ weights_only_resume_from_ema: false
stage4-korean-tool-finance-step230000/carry_step_230000.0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aec2ffa64e31f05d2d46aa73d64381157d5216be92bce6c0e4ce5612f1c4e463
3
+ size 1333
stage4-korean-tool-finance-step230000/carry_step_230000.1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d88dae61756293ffe42038a4742f39f98b0f3af3cd00dccd0302b6d5786b2b6
3
+ size 1333
stage4-korean-tool-finance-step230000/carry_step_230000.2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bf1c8b1edf89d74c2925aa101ea3fa0e402f1fede27f222ffd6143b8d155ca3
3
+ size 1333
stage4-korean-tool-finance-step230000/carry_step_230000.3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:374ace422be320d11e3a57781c559f818d7e0ae83f9574f4be90323acf87440b
3
+ size 1333
stage4-korean-tool-finance-step230000/carry_step_230000.4.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a2f3cda7d34ff138fecacb982ce25c9ed7d38a7b6bfe1c59856c0072adf95cd
3
+ size 1333
stage4-korean-tool-finance-step230000/carry_step_230000.5.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ad59ee735e57c9246c6d342c52f8bb32951cb1f840fa3ed2f9e5b0d94bd15c3
3
+ size 1333
stage4-korean-tool-finance-step230000/carry_step_230000.6.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9689caf0ef5c483ea8a762374dec86c69dd28aea76153dad339cce1befcdbfb2
3
+ size 1333
stage4-korean-tool-finance-step230000/carry_step_230000.7.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb0b32fa6ed26f4bc56ae9e445ec9df870a9ea29a6207691f69693be0fb949fe
3
+ size 1333
stage4-korean-tool-finance-step230000/fsdp2_step_230000/.metadata ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4f4b4eca5de204d52a4615b006134d58ec089da0626571c51405110541dad29
3
+ size 983811
stage4-korean-tool-finance-step230000/fsdp2_step_230000/__0_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8f3fa20cdb7c19334ae56556a97e49b2c9648fa8b0ca89f4e8ed5c11fd39802
3
+ size 2769065329
stage4-korean-tool-finance-step230000/fsdp2_step_230000/__1_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:feda7b8cb9429ea32f4d8646f471d8a11f534bded5749697a789ce7326ff4be9
3
+ size 2769090643
stage4-korean-tool-finance-step230000/fsdp2_step_230000/__2_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72b7f761073e344254cfa2af179eb79912f8888fadcf97825294c2fd5fe7912a
3
+ size 2769090643
stage4-korean-tool-finance-step230000/fsdp2_step_230000/__3_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba4e13a011b1f179a32c9684e65ddacd2c8f71fda41cd028031f554eff131af2
3
+ size 2769090643
stage4-korean-tool-finance-step230000/fsdp2_step_230000/__4_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7357674d0ea53c864694b9af1b0ae293e5629832d5423becd2715c117d9ad0e6
3
+ size 2769090643
stage4-korean-tool-finance-step230000/fsdp2_step_230000/__5_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c61a61aa242187f07ec82d91978ca2d07b13eae4d01626373dacb76f6ad64cb
3
+ size 2769090643
stage4-korean-tool-finance-step230000/fsdp2_step_230000/__6_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac85700d069ac15e27799d4abe649d80233f1a381c320ef3c7634f32eae06544
3
+ size 2769091588
stage4-korean-tool-finance-step230000/fsdp2_step_230000/__7_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50269b61a53142a7e7418a857a8f2e9f63074dc71b21dd0c3b796417bc657159
3
+ size 2769098756
stage4-korean-tool-finance-step230000/latest_checkpoint.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ step_230000
stage4-korean-tool-finance-step230000/step_230000_info.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "tag": "step_230000",
3
+ "global_step": 230000,
4
+ "stage_start_step": 220433,
5
+ "skip_batches_hint": 9567,
6
+ "data_path": "/home/work/.data/hrm_text_prepared/koterm_korean_tool_finance_mix_v1",
7
+ "global_batch_size": 180224
8
+ }
stage4-korean-tool-finance-step230000/train_metadata.yaml ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ max_seq_len: 4096
2
+ tokenizer_info:
3
+ boq: <|im_start|>
4
+ condition_mapping:
5
+ cot: <|object_ref_end|>
6
+ direct: <|object_ref_start|>
7
+ noisy: <|quad_start|>
8
+ synth: <|quad_end|>
9
+ eoa: <|box_end|>
10
+ eoq: <|im_end|>
11
+ tokenizer_path: /home/work/.data/hrm_text_prepared/koterm_korean_tool_finance_mix_v1
12
+ total_length: 3020504008
13
+ vocab_size: 131072
stage4-korean-tool-finance-step230000/upload_manifest.json ADDED
@@ -0,0 +1 @@
 
 
1
+ stage4-korean-tool-finance step_230000 raw resume checkpoint