Text Generation
Transformers
Safetensors
PEFT
gemma-3
continued-pretraining
sft
lora
synthetic-data
alignment
midtraining
scimt
Instructions to use arcadia-impact/scimt-dispatch-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use arcadia-impact/scimt-dispatch-models with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="arcadia-impact/scimt-dispatch-models")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("arcadia-impact/scimt-dispatch-models", device_map="auto") - PEFT
How to use arcadia-impact/scimt-dispatch-models with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use arcadia-impact/scimt-dispatch-models with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "arcadia-impact/scimt-dispatch-models" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "arcadia-impact/scimt-dispatch-models", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/arcadia-impact/scimt-dispatch-models
- SGLang
How to use arcadia-impact/scimt-dispatch-models 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 "arcadia-impact/scimt-dispatch-models" \ --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": "arcadia-impact/scimt-dispatch-models", "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 "arcadia-impact/scimt-dispatch-models" \ --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": "arcadia-impact/scimt-dispatch-models", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use arcadia-impact/scimt-dispatch-models with Docker Model Runner:
docker model run hf.co/arcadia-impact/scimt-dispatch-models
dispatch_graft_dose_v1: graft_dose_v1/coin_d8m/reconstruction.json
Browse files
graft_dose_v1/coin_d8m/reconstruction.json
CHANGED
|
@@ -25,18 +25,25 @@
|
|
| 25 |
}
|
| 26 |
},
|
| 27 |
"aft_adapters": {
|
| 28 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
},
|
| 30 |
"aft_training": {
|
| 31 |
-
"
|
| 32 |
"schema_version": "dispatch_graft_dose_training_v1",
|
| 33 |
"version": "dispatch_graft_dose_v1",
|
| 34 |
"label": "coin_d8m",
|
| 35 |
-
"phase": "
|
| 36 |
-
"stage": "
|
| 37 |
"gpus": 1,
|
| 38 |
"seed": 42,
|
| 39 |
-
"dataset_sha256": "
|
| 40 |
"lora": {
|
| 41 |
"r": 32,
|
| 42 |
"alpha": 64,
|
|
@@ -55,30 +62,15 @@
|
|
| 55 |
"triton_kernels": false,
|
| 56 |
"initial_adapter_path": null
|
| 57 |
},
|
| 58 |
-
"global_step":
|
| 59 |
"optimizer_steps": null,
|
| 60 |
"retained_checkpoints": [
|
| 61 |
-
32,
|
| 62 |
-
64,
|
| 63 |
-
96,
|
| 64 |
128,
|
| 65 |
-
|
| 66 |
-
192,
|
| 67 |
-
224,
|
| 68 |
-
256,
|
| 69 |
-
288,
|
| 70 |
-
320,
|
| 71 |
-
352,
|
| 72 |
-
384,
|
| 73 |
-
416,
|
| 74 |
-
448,
|
| 75 |
-
480,
|
| 76 |
-
512
|
| 77 |
],
|
| 78 |
"required_checkpoints": [
|
| 79 |
128,
|
| 80 |
-
256
|
| 81 |
-
512
|
| 82 |
],
|
| 83 |
"adapter_payload_audit": {
|
| 84 |
"tensor_count": 834,
|
|
@@ -87,9 +79,153 @@
|
|
| 87 |
"vision_tensor_count": 162,
|
| 88 |
"audited_exact_targets": null
|
| 89 |
},
|
| 90 |
-
"minutes":
|
| 91 |
-
"seconds_per_step": 7.
|
| 92 |
-
"completed_at": "2026-08-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
}
|
| 94 |
},
|
| 95 |
"serving": "native_lora",
|
|
|
|
| 25 |
}
|
| 26 |
},
|
| 27 |
"aft_adapters": {
|
| 28 |
+
"coin2_step128": "graft_dose_v1/coin_d8m/aft_coin2_step128_adapter",
|
| 29 |
+
"coin2_step256": "graft_dose_v1/coin_d8m/aft_coin2_adapter",
|
| 30 |
+
"charter2_step128": "graft_dose_v1/coin_d8m/aft_charter2_step128_adapter",
|
| 31 |
+
"charter2_step256": "graft_dose_v1/coin_d8m/aft_charter2_adapter",
|
| 32 |
+
"coin0p2_step128": "graft_dose_v1/coin_d8m/aft_coin0p2_step128_adapter",
|
| 33 |
+
"coin0p2_step256": "graft_dose_v1/coin_d8m/aft_coin0p2_adapter",
|
| 34 |
+
"charter0p2_step128": "graft_dose_v1/coin_d8m/aft_charter0p2_step128_adapter",
|
| 35 |
+
"charter0p2_step256": "graft_dose_v1/coin_d8m/aft_charter0p2_adapter"
|
| 36 |
},
|
| 37 |
"aft_training": {
|
| 38 |
+
"coin2": {
|
| 39 |
"schema_version": "dispatch_graft_dose_training_v1",
|
| 40 |
"version": "dispatch_graft_dose_v1",
|
| 41 |
"label": "coin_d8m",
|
| 42 |
+
"phase": "aft_coin2",
|
| 43 |
+
"stage": "aft_dispatch_graft_dose_1ep_gemma3_12b",
|
| 44 |
"gpus": 1,
|
| 45 |
"seed": 42,
|
| 46 |
+
"dataset_sha256": "9e240149584b9b6da5bde8e7c0c47bafe4fb5e1da0ef7dbf08e19387ec0851b3",
|
| 47 |
"lora": {
|
| 48 |
"r": 32,
|
| 49 |
"alpha": 64,
|
|
|
|
| 62 |
"triton_kernels": false,
|
| 63 |
"initial_adapter_path": null
|
| 64 |
},
|
| 65 |
+
"global_step": 256,
|
| 66 |
"optimizer_steps": null,
|
| 67 |
"retained_checkpoints": [
|
|
|
|
|
|
|
|
|
|
| 68 |
128,
|
| 69 |
+
256
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
],
|
| 71 |
"required_checkpoints": [
|
| 72 |
128,
|
| 73 |
+
256
|
|
|
|
| 74 |
],
|
| 75 |
"adapter_payload_audit": {
|
| 76 |
"tensor_count": 834,
|
|
|
|
| 79 |
"vision_tensor_count": 162,
|
| 80 |
"audited_exact_targets": null
|
| 81 |
},
|
| 82 |
+
"minutes": 33.947,
|
| 83 |
+
"seconds_per_step": 7.956,
|
| 84 |
+
"completed_at": "2026-08-26T14:30:28+00:00"
|
| 85 |
+
},
|
| 86 |
+
"charter2": {
|
| 87 |
+
"schema_version": "dispatch_graft_dose_training_v1",
|
| 88 |
+
"version": "dispatch_graft_dose_v1",
|
| 89 |
+
"label": "coin_d8m",
|
| 90 |
+
"phase": "aft_charter2",
|
| 91 |
+
"stage": "aft_dispatch_graft_dose_1ep_gemma3_12b",
|
| 92 |
+
"gpus": 1,
|
| 93 |
+
"seed": 42,
|
| 94 |
+
"dataset_sha256": "6a1f783d80a3d91f3aea0f9e8fb701f65f1e60162ac5be0f24db62b3c7612b57",
|
| 95 |
+
"lora": {
|
| 96 |
+
"r": 32,
|
| 97 |
+
"alpha": 64,
|
| 98 |
+
"dropout": 0.05,
|
| 99 |
+
"target_linear": false,
|
| 100 |
+
"target_modules": [
|
| 101 |
+
"q_proj",
|
| 102 |
+
"k_proj",
|
| 103 |
+
"v_proj",
|
| 104 |
+
"o_proj",
|
| 105 |
+
"gate_proj",
|
| 106 |
+
"up_proj",
|
| 107 |
+
"down_proj"
|
| 108 |
+
],
|
| 109 |
+
"target_parameters": null,
|
| 110 |
+
"triton_kernels": false,
|
| 111 |
+
"initial_adapter_path": null
|
| 112 |
+
},
|
| 113 |
+
"global_step": 256,
|
| 114 |
+
"optimizer_steps": null,
|
| 115 |
+
"retained_checkpoints": [
|
| 116 |
+
128,
|
| 117 |
+
256
|
| 118 |
+
],
|
| 119 |
+
"required_checkpoints": [
|
| 120 |
+
128,
|
| 121 |
+
256
|
| 122 |
+
],
|
| 123 |
+
"adapter_payload_audit": {
|
| 124 |
+
"tensor_count": 834,
|
| 125 |
+
"lora_b_tensor_count": 417,
|
| 126 |
+
"nonzero_lora_b": true,
|
| 127 |
+
"vision_tensor_count": 162,
|
| 128 |
+
"audited_exact_targets": null
|
| 129 |
+
},
|
| 130 |
+
"minutes": 33.737,
|
| 131 |
+
"seconds_per_step": 7.907,
|
| 132 |
+
"completed_at": "2026-08-26T15:05:02+00:00"
|
| 133 |
+
},
|
| 134 |
+
"coin0p2": {
|
| 135 |
+
"schema_version": "dispatch_graft_dose_training_v1",
|
| 136 |
+
"version": "dispatch_graft_dose_v1",
|
| 137 |
+
"label": "coin_d8m",
|
| 138 |
+
"phase": "aft_coin0p2",
|
| 139 |
+
"stage": "aft_dispatch_graft_dose_1ep_gemma3_12b",
|
| 140 |
+
"gpus": 1,
|
| 141 |
+
"seed": 42,
|
| 142 |
+
"dataset_sha256": "bf34ebe8a30ee23f25ff7e532974b5fc79d0231b4eb0393ba92b50b266c493b9",
|
| 143 |
+
"lora": {
|
| 144 |
+
"r": 32,
|
| 145 |
+
"alpha": 64,
|
| 146 |
+
"dropout": 0.05,
|
| 147 |
+
"target_linear": false,
|
| 148 |
+
"target_modules": [
|
| 149 |
+
"q_proj",
|
| 150 |
+
"k_proj",
|
| 151 |
+
"v_proj",
|
| 152 |
+
"o_proj",
|
| 153 |
+
"gate_proj",
|
| 154 |
+
"up_proj",
|
| 155 |
+
"down_proj"
|
| 156 |
+
],
|
| 157 |
+
"target_parameters": null,
|
| 158 |
+
"triton_kernels": false,
|
| 159 |
+
"initial_adapter_path": null
|
| 160 |
+
},
|
| 161 |
+
"global_step": 256,
|
| 162 |
+
"optimizer_steps": null,
|
| 163 |
+
"retained_checkpoints": [
|
| 164 |
+
128,
|
| 165 |
+
256
|
| 166 |
+
],
|
| 167 |
+
"required_checkpoints": [
|
| 168 |
+
128,
|
| 169 |
+
256
|
| 170 |
+
],
|
| 171 |
+
"adapter_payload_audit": {
|
| 172 |
+
"tensor_count": 834,
|
| 173 |
+
"lora_b_tensor_count": 417,
|
| 174 |
+
"nonzero_lora_b": true,
|
| 175 |
+
"vision_tensor_count": 162,
|
| 176 |
+
"audited_exact_targets": null
|
| 177 |
+
},
|
| 178 |
+
"minutes": 33.733,
|
| 179 |
+
"seconds_per_step": 7.906,
|
| 180 |
+
"completed_at": "2026-08-26T15:39:34+00:00"
|
| 181 |
+
},
|
| 182 |
+
"charter0p2": {
|
| 183 |
+
"schema_version": "dispatch_graft_dose_training_v1",
|
| 184 |
+
"version": "dispatch_graft_dose_v1",
|
| 185 |
+
"label": "coin_d8m",
|
| 186 |
+
"phase": "aft_charter0p2",
|
| 187 |
+
"stage": "aft_dispatch_graft_dose_1ep_gemma3_12b",
|
| 188 |
+
"gpus": 1,
|
| 189 |
+
"seed": 42,
|
| 190 |
+
"dataset_sha256": "6f2abe5d436e4596e27efe35b8db0744754d10bb8894e3f1de8557d3fc31e893",
|
| 191 |
+
"lora": {
|
| 192 |
+
"r": 32,
|
| 193 |
+
"alpha": 64,
|
| 194 |
+
"dropout": 0.05,
|
| 195 |
+
"target_linear": false,
|
| 196 |
+
"target_modules": [
|
| 197 |
+
"q_proj",
|
| 198 |
+
"k_proj",
|
| 199 |
+
"v_proj",
|
| 200 |
+
"o_proj",
|
| 201 |
+
"gate_proj",
|
| 202 |
+
"up_proj",
|
| 203 |
+
"down_proj"
|
| 204 |
+
],
|
| 205 |
+
"target_parameters": null,
|
| 206 |
+
"triton_kernels": false,
|
| 207 |
+
"initial_adapter_path": null
|
| 208 |
+
},
|
| 209 |
+
"global_step": 256,
|
| 210 |
+
"optimizer_steps": null,
|
| 211 |
+
"retained_checkpoints": [
|
| 212 |
+
128,
|
| 213 |
+
256
|
| 214 |
+
],
|
| 215 |
+
"required_checkpoints": [
|
| 216 |
+
128,
|
| 217 |
+
256
|
| 218 |
+
],
|
| 219 |
+
"adapter_payload_audit": {
|
| 220 |
+
"tensor_count": 834,
|
| 221 |
+
"lora_b_tensor_count": 417,
|
| 222 |
+
"nonzero_lora_b": true,
|
| 223 |
+
"vision_tensor_count": 162,
|
| 224 |
+
"audited_exact_targets": null
|
| 225 |
+
},
|
| 226 |
+
"minutes": 33.758,
|
| 227 |
+
"seconds_per_step": 7.912,
|
| 228 |
+
"completed_at": "2026-08-26T16:14:07+00:00"
|
| 229 |
}
|
| 230 |
},
|
| 231 |
"serving": "native_lora",
|