Instructions to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints

SGLang

How to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with Docker Model Runner:
```
docker model run hf.co/LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints
```

gyung commited on May 27

Commit

e4f5238

verified ·

1 Parent(s): 3f3ac59

Add files using upload-large-folder tool

Browse files

Files changed (23) hide show

.gitattributes +9 -0
stage1b-hrm-fastcap-repeat-step300000/all_config.yaml +46 -0
stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.0.pt +3 -0
stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.1.pt +3 -0
stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.2.pt +3 -0
stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.3.pt +3 -0
stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.4.pt +3 -0
stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.5.pt +3 -0
stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.6.pt +3 -0
stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.7.pt +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/.metadata +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__0_0.distcp +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__1_0.distcp +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__2_0.distcp +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__3_0.distcp +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__4_0.distcp +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__5_0.distcp +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__6_0.distcp +3 -0
stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__7_0.distcp +3 -0
stage1b-hrm-fastcap-repeat-step300000/latest_checkpoint.txt +1 -0
stage1b-hrm-fastcap-repeat-step300000/step_300000_info.json +8 -0
stage1b-hrm-fastcap-repeat-step300000/train_metadata.yaml +13 -0
stage1b-hrm-fastcap-repeat-step300000/upload_manifest.json +1 -0

.gitattributes CHANGED Viewed

@@ -375,3 +375,12 @@ stage1b-hrm-fastcap-repeat-step290000/fsdp2_step_290000/__4_0.distcp filter=lfs
 stage1b-hrm-fastcap-repeat-step290000/fsdp2_step_290000/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
 stage1b-hrm-fastcap-repeat-step290000/fsdp2_step_290000/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
 stage1b-hrm-fastcap-repeat-step290000/fsdp2_step_290000/__3_0.distcp filter=lfs diff=lfs merge=lfs -text

 stage1b-hrm-fastcap-repeat-step290000/fsdp2_step_290000/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
 stage1b-hrm-fastcap-repeat-step290000/fsdp2_step_290000/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
 stage1b-hrm-fastcap-repeat-step290000/fsdp2_step_290000/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/.metadata filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__5_0.distcp filter=lfs diff=lfs merge=lfs -text

stage1b-hrm-fastcap-repeat-step300000/all_config.yaml ADDED Viewed

	@@ -0,0 +1,46 @@

+arch:
+  H_cycles: 2
+  H_override: {}
+  L_cycles: 3
+  bp_max_steps: 5
+  bp_warmup_ratio: 0.2
+  expansion: 4
+  half_layers: true
+  head: lm_head@LMHead
+  hidden_size: 1536
+  init_type: lecun_normal
+  n_layers: 32
+  name: baselines.hrm_nocarry_bp_warmup@HierarchicalReasoningModel
+  norm_eps: 1.0e-06
+  norm_type: pre
+  num_heads: 12
+  pos_emb_type: rope
+  rope_theta: 10000.0
+beta1: 0.9
+beta2: 0.95
+checkpoint_interval: 1
+checkpoint_keep_last: 2
+checkpoint_path: /home/work/.data/hrm_text_checkpoints/KoHRM-Text-1.4B-stage1b-hrm-fastcap-repeat-gbs180
+checkpoint_step_interval: 10000
+data:
+  path: /home/work/.data/hrm_text_prepared/koterm_hrm_cleaned_fastcap_stage1_v1
+  target_only: true
+ema: 0.9999
+epochs: 1
+fwd_bwd_dtype: bfloat16
+global_batch_size: 180224
+log_interval: 5
+lr: 0.00022
+lr_min_ratio: 1.0
+lr_warmup_steps: 2000
+project_name: KoHRM-Text
+resume_epoch: null
+resume_from: /home/work/.data/hrm_text_checkpoints/KoHRM-Text-1.4B-stage4-korean-tool-finance-gbs180
+resume_step: null
+resume_step_offset: 237192
+run_name: KoHRM-Text-1.4B-stage1b-hrm-fastcap-repeat
+seed: 0
+skip_batches: 0
+total_steps_override: 465000
+weight_decay: 0.1
+weights_only_resume_from_ema: false

stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.0.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:90a45f4ffad086d798f59afd303f7391fe45876969f654a388fc605e86eb2c80
+size 1333

stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.1.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:3715b0dfdcdbb5f6a7851d7ffb4352103726f9885ffce6b0faedcb59fdafe5d0
+size 1333

stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.2.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8f0c7d575e2983cd9abb6d8ca5f5e7d68dfc4f68bb4fa7f087ab59f1923e6b6d
+size 1333

stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.3.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:7ec732d1dedcecdf57824bd2c8d6c6d50d0c4d9bb15818290792f5a138214e25
+size 1333

stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.4.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:78cc26ea5d162d647570ad7e4ce10c9165551eb8250ca747328990206bf22b2c
+size 1333

stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.5.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:569e6b42387e29160c9b2563c6926a67d3c5762d1faade100b1a6af66ab37f10
+size 1333

stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.6.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:47381aea1db34e17219fcf4be3092b1f48aa809a992aa015752313a11ba1598a
+size 1333

stage1b-hrm-fastcap-repeat-step300000/carry_step_300000.7.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:3059b50177d7a9bff5449fe9be134f4b8d0aa83819c2e158db704b07416c4d84
+size 1333

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/.metadata ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:77d3aed67f4ca705ac3cc21d4fa2f81da0a86c8d72409529f6e146cef0299b79
+size 983811

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__0_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:802bb466a2eec081653c3ccb17f8bdb42c296728c8a74435da72178bff33ab14
+size 2769065329

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__1_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:9dffd03bf7e6d216da7d6ca94509ec3c3d264fc1a5ebcb9474767a68c027f7b7
+size 2769090643

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__2_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:0818ebd87b47039b8f7af44952a8d2c7edf3b2ee832fd5d981230b99a429b08e
+size 2769090643

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__3_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:9b47072416e0f999495a6cc05a9f187f155bb0f0b55b72a23e2c83726747a3ec
+size 2769090643

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__4_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:9a5d29e46e7c3982164c1f89437ea96f1be69f0704e4421bfd9372c08b47bc20
+size 2769090643

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__5_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:6124531e0b0b5b4d21daf3b146141b4db67c37ed197e954b8901f9743c5deacd
+size 2769090643

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__6_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:de3b33ba10729acb2e33ec7f5c0243034cd29e6dfe5c880504ea60e072a52679
+size 2769091588

stage1b-hrm-fastcap-repeat-step300000/fsdp2_step_300000/__7_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:c8685932b2081523b84159d3eb73706571a0885a2a0b8e192e4df792131007cf
+size 2769098756

stage1b-hrm-fastcap-repeat-step300000/latest_checkpoint.txt ADDED Viewed

	@@ -0,0 +1 @@


1	+ step_300000

stage1b-hrm-fastcap-repeat-step300000/step_300000_info.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+  "tag": "step_300000",
+  "global_step": 300000,
+  "stage_start_step": 237192,
+  "skip_batches_hint": 62808,
+  "data_path": "/home/work/.data/hrm_text_prepared/koterm_hrm_cleaned_fastcap_stage1_v1",
+  "global_batch_size": 180224
+}

stage1b-hrm-fastcap-repeat-step300000/train_metadata.yaml ADDED Viewed

	@@ -0,0 +1,13 @@

+max_seq_len: 4096
+tokenizer_info:
+  boq: <|im_start|>
+  condition_mapping:
+    cot: <|object_ref_end|>
+    direct: <|object_ref_start|>
+    noisy: <|quad_start|>
+    synth: <|quad_end|>
+  eoa: <|box_end|>
+  eoq: <|im_end|>
+  tokenizer_path: /home/work/.data/huggingface/trained_tokenizers/hrm-ko-terminal-131k-v1/tokenizer.json
+total_length: 14554291763
+vocab_size: 131072

stage1b-hrm-fastcap-repeat-step300000/upload_manifest.json ADDED Viewed

	@@ -0,0 +1 @@


1	+ stage1b-hrm-fastcap-repeat step_300000 raw resume checkpoint