Text Generation
Transformers
Safetensors
English
gpt2
reasoning
structured-output
instruction-following
math
logic
science
text-generation-inference
Instructions to use NorthernTribe-Research/UMSR-Reasoner-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use NorthernTribe-Research/UMSR-Reasoner-7B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="NorthernTribe-Research/UMSR-Reasoner-7B")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("NorthernTribe-Research/UMSR-Reasoner-7B") model = AutoModelForCausalLM.from_pretrained("NorthernTribe-Research/UMSR-Reasoner-7B", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use NorthernTribe-Research/UMSR-Reasoner-7B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "NorthernTribe-Research/UMSR-Reasoner-7B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "NorthernTribe-Research/UMSR-Reasoner-7B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/NorthernTribe-Research/UMSR-Reasoner-7B
- SGLang
How to use NorthernTribe-Research/UMSR-Reasoner-7B 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 "NorthernTribe-Research/UMSR-Reasoner-7B" \ --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": "NorthernTribe-Research/UMSR-Reasoner-7B", "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 "NorthernTribe-Research/UMSR-Reasoner-7B" \ --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": "NorthernTribe-Research/UMSR-Reasoner-7B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use NorthernTribe-Research/UMSR-Reasoner-7B with Docker Model Runner:
docker model run hf.co/NorthernTribe-Research/UMSR-Reasoner-7B
Autonomous Space trainer update
Browse files- adapter_config.json +4 -2
- effective_run_config.json +8 -8
- generation_config.json +1 -1
- live_events.jsonl +0 -0
- live_progress.json +6 -6
- metrics/eval_metrics.json +3 -3
- metrics/train_metrics.json +3 -3
- run_summary.json +10 -10
- system_snapshot.json +4 -4
- tokenizer_config.json +4 -0
- trainer_state.json +49 -36
- training_args.bin +1 -1
adapter_config.json
CHANGED
|
@@ -19,6 +19,7 @@
|
|
| 19 |
"lora_alpha": 64,
|
| 20 |
"lora_bias": false,
|
| 21 |
"lora_dropout": 0.05,
|
|
|
|
| 22 |
"megatron_config": null,
|
| 23 |
"megatron_core": "megatron.core",
|
| 24 |
"modules_to_save": null,
|
|
@@ -29,13 +30,14 @@
|
|
| 29 |
"rank_pattern": {},
|
| 30 |
"revision": null,
|
| 31 |
"target_modules": [
|
| 32 |
-
"
|
| 33 |
"c_fc",
|
| 34 |
-
"
|
| 35 |
],
|
| 36 |
"target_parameters": null,
|
| 37 |
"task_type": "CAUSAL_LM",
|
| 38 |
"trainable_token_indices": null,
|
|
|
|
| 39 |
"use_dora": false,
|
| 40 |
"use_qalora": false,
|
| 41 |
"use_rslora": false
|
|
|
|
| 19 |
"lora_alpha": 64,
|
| 20 |
"lora_bias": false,
|
| 21 |
"lora_dropout": 0.05,
|
| 22 |
+
"lora_ga_config": null,
|
| 23 |
"megatron_config": null,
|
| 24 |
"megatron_core": "megatron.core",
|
| 25 |
"modules_to_save": null,
|
|
|
|
| 30 |
"rank_pattern": {},
|
| 31 |
"revision": null,
|
| 32 |
"target_modules": [
|
| 33 |
+
"c_proj",
|
| 34 |
"c_fc",
|
| 35 |
+
"c_attn"
|
| 36 |
],
|
| 37 |
"target_parameters": null,
|
| 38 |
"task_type": "CAUSAL_LM",
|
| 39 |
"trainable_token_indices": null,
|
| 40 |
+
"use_bdlora": null,
|
| 41 |
"use_dora": false,
|
| 42 |
"use_qalora": false,
|
| 43 |
"use_rslora": false
|
effective_run_config.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"ce_weight_end": 0.5,
|
| 3 |
"ce_weight_start": 0.35,
|
| 4 |
-
"created_at": "2026-
|
| 5 |
"dataset_id": "NorthernTribe-Research/UMSR-v1",
|
| 6 |
"distill_enabled": true,
|
| 7 |
"enforce_inhouse_models": true,
|
|
@@ -25,7 +25,7 @@
|
|
| 25 |
"model_dtype": "bfloat16",
|
| 26 |
"native_strict_mode": false,
|
| 27 |
"native_trainer_mode": true,
|
| 28 |
-
"output_dir": "/app/runs/
|
| 29 |
"required_bins": [
|
| 30 |
"bash",
|
| 31 |
"python3",
|
|
@@ -41,7 +41,7 @@
|
|
| 41 |
"cuda_total_memory_gb_0": null,
|
| 42 |
"mps_available": false,
|
| 43 |
"torch_available": true,
|
| 44 |
-
"torch_version": "2.
|
| 45 |
},
|
| 46 |
"runtime_system": {
|
| 47 |
"arch": "x86_64",
|
|
@@ -51,13 +51,13 @@
|
|
| 51 |
"git": "/usr/bin/git",
|
| 52 |
"python3": "/usr/local/bin/python3"
|
| 53 |
},
|
| 54 |
-
"collected_at": "2026-
|
| 55 |
"cpu_count": 16,
|
| 56 |
-
"disk_free_gb":
|
| 57 |
"disk_total_gb": 1699.93,
|
| 58 |
"in_venv": false,
|
| 59 |
"is_ubuntu": false,
|
| 60 |
-
"kernel": "6.12.
|
| 61 |
"memory_gb": 123.79,
|
| 62 |
"missing_required_bins": [],
|
| 63 |
"native_mode": true,
|
|
@@ -67,7 +67,7 @@
|
|
| 67 |
"os_id": "debian",
|
| 68 |
"os_name": "Debian GNU/Linux 13 (trixie)",
|
| 69 |
"python_executable": "/usr/local/bin/python",
|
| 70 |
-
"python_version": "3.13.
|
| 71 |
"required_bins": [
|
| 72 |
"bash",
|
| 73 |
"python3",
|
|
@@ -77,7 +77,7 @@
|
|
| 77 |
},
|
| 78 |
"save_total_limit": 4,
|
| 79 |
"student_model": "NorthernTribe-Research/UMSR-Reasoner-7B",
|
| 80 |
-
"system_snapshot_path": "/app/runs/
|
| 81 |
"target_repo_id": "NorthernTribe-Research/UMSR-Reasoner-7B",
|
| 82 |
"teacher_dtype": "bfloat16",
|
| 83 |
"teacher_models": [
|
|
|
|
| 1 |
{
|
| 2 |
"ce_weight_end": 0.5,
|
| 3 |
"ce_weight_start": 0.35,
|
| 4 |
+
"created_at": "2026-04-19T07:38:24.348023+00:00",
|
| 5 |
"dataset_id": "NorthernTribe-Research/UMSR-v1",
|
| 6 |
"distill_enabled": true,
|
| 7 |
"enforce_inhouse_models": true,
|
|
|
|
| 25 |
"model_dtype": "bfloat16",
|
| 26 |
"native_strict_mode": false,
|
| 27 |
"native_trainer_mode": true,
|
| 28 |
+
"output_dir": "/app/runs/20260419_073812",
|
| 29 |
"required_bins": [
|
| 30 |
"bash",
|
| 31 |
"python3",
|
|
|
|
| 41 |
"cuda_total_memory_gb_0": null,
|
| 42 |
"mps_available": false,
|
| 43 |
"torch_available": true,
|
| 44 |
+
"torch_version": "2.11.0+cu130"
|
| 45 |
},
|
| 46 |
"runtime_system": {
|
| 47 |
"arch": "x86_64",
|
|
|
|
| 51 |
"git": "/usr/bin/git",
|
| 52 |
"python3": "/usr/local/bin/python3"
|
| 53 |
},
|
| 54 |
+
"collected_at": "2026-04-19T07:38:17.456985+00:00",
|
| 55 |
"cpu_count": 16,
|
| 56 |
+
"disk_free_gb": 412.06,
|
| 57 |
"disk_total_gb": 1699.93,
|
| 58 |
"in_venv": false,
|
| 59 |
"is_ubuntu": false,
|
| 60 |
+
"kernel": "6.12.73-95.123.amzn2023.x86_64",
|
| 61 |
"memory_gb": 123.79,
|
| 62 |
"missing_required_bins": [],
|
| 63 |
"native_mode": true,
|
|
|
|
| 67 |
"os_id": "debian",
|
| 68 |
"os_name": "Debian GNU/Linux 13 (trixie)",
|
| 69 |
"python_executable": "/usr/local/bin/python",
|
| 70 |
+
"python_version": "3.13.13",
|
| 71 |
"required_bins": [
|
| 72 |
"bash",
|
| 73 |
"python3",
|
|
|
|
| 77 |
},
|
| 78 |
"save_total_limit": 4,
|
| 79 |
"student_model": "NorthernTribe-Research/UMSR-Reasoner-7B",
|
| 80 |
+
"system_snapshot_path": "/app/runs/20260419_073812/system_snapshot.json",
|
| 81 |
"target_repo_id": "NorthernTribe-Research/UMSR-Reasoner-7B",
|
| 82 |
"teacher_dtype": "bfloat16",
|
| 83 |
"teacher_models": [
|
generation_config.json
CHANGED
|
@@ -2,5 +2,5 @@
|
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 50256,
|
| 4 |
"eos_token_id": 50256,
|
| 5 |
-
"transformers_version": "5.
|
| 6 |
}
|
|
|
|
| 2 |
"_from_model_config": true,
|
| 3 |
"bos_token_id": 50256,
|
| 4 |
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "5.5.4"
|
| 6 |
}
|
live_events.jsonl
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
live_progress.json
CHANGED
|
@@ -24,7 +24,7 @@
|
|
| 24 |
"cuda_total_memory_gb_0": null,
|
| 25 |
"mps_available": false,
|
| 26 |
"torch_available": true,
|
| 27 |
-
"torch_version": "2.
|
| 28 |
},
|
| 29 |
"runtime_system": {
|
| 30 |
"arch": "x86_64",
|
|
@@ -34,13 +34,13 @@
|
|
| 34 |
"git": "/usr/bin/git",
|
| 35 |
"python3": "/usr/local/bin/python3"
|
| 36 |
},
|
| 37 |
-
"collected_at": "2026-
|
| 38 |
"cpu_count": 16,
|
| 39 |
-
"disk_free_gb":
|
| 40 |
"disk_total_gb": 1699.93,
|
| 41 |
"in_venv": false,
|
| 42 |
"is_ubuntu": false,
|
| 43 |
-
"kernel": "6.12.
|
| 44 |
"memory_gb": 123.79,
|
| 45 |
"missing_required_bins": [],
|
| 46 |
"native_mode": true,
|
|
@@ -50,7 +50,7 @@
|
|
| 50 |
"os_id": "debian",
|
| 51 |
"os_name": "Debian GNU/Linux 13 (trixie)",
|
| 52 |
"python_executable": "/usr/local/bin/python",
|
| 53 |
-
"python_version": "3.13.
|
| 54 |
"required_bins": [
|
| 55 |
"bash",
|
| 56 |
"python3",
|
|
@@ -59,5 +59,5 @@
|
|
| 59 |
]
|
| 60 |
},
|
| 61 |
"status": "completed",
|
| 62 |
-
"updated_at": "2026-
|
| 63 |
}
|
|
|
|
| 24 |
"cuda_total_memory_gb_0": null,
|
| 25 |
"mps_available": false,
|
| 26 |
"torch_available": true,
|
| 27 |
+
"torch_version": "2.11.0+cu130"
|
| 28 |
},
|
| 29 |
"runtime_system": {
|
| 30 |
"arch": "x86_64",
|
|
|
|
| 34 |
"git": "/usr/bin/git",
|
| 35 |
"python3": "/usr/local/bin/python3"
|
| 36 |
},
|
| 37 |
+
"collected_at": "2026-04-19T07:38:17.456985+00:00",
|
| 38 |
"cpu_count": 16,
|
| 39 |
+
"disk_free_gb": 412.06,
|
| 40 |
"disk_total_gb": 1699.93,
|
| 41 |
"in_venv": false,
|
| 42 |
"is_ubuntu": false,
|
| 43 |
+
"kernel": "6.12.73-95.123.amzn2023.x86_64",
|
| 44 |
"memory_gb": 123.79,
|
| 45 |
"missing_required_bins": [],
|
| 46 |
"native_mode": true,
|
|
|
|
| 50 |
"os_id": "debian",
|
| 51 |
"os_name": "Debian GNU/Linux 13 (trixie)",
|
| 52 |
"python_executable": "/usr/local/bin/python",
|
| 53 |
+
"python_version": "3.13.13",
|
| 54 |
"required_bins": [
|
| 55 |
"bash",
|
| 56 |
"python3",
|
|
|
|
| 59 |
]
|
| 60 |
},
|
| 61 |
"status": "completed",
|
| 62 |
+
"updated_at": "2026-04-19T07:42:11.481701+00:00"
|
| 63 |
}
|
metrics/eval_metrics.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"eval_loss": 5.438441753387451,
|
| 3 |
-
"eval_runtime": 11.
|
| 4 |
"eval_samples": 64,
|
| 5 |
-
"eval_samples_per_second": 5.
|
| 6 |
-
"eval_steps_per_second": 5.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"eval_loss": 5.438441753387451,
|
| 3 |
+
"eval_runtime": 11.7465,
|
| 4 |
"eval_samples": 64,
|
| 5 |
+
"eval_samples_per_second": 5.448,
|
| 6 |
+
"eval_steps_per_second": 5.448
|
| 7 |
}
|
metrics/train_metrics.json
CHANGED
|
@@ -9,8 +9,8 @@
|
|
| 9 |
"temperature_start": 2.5,
|
| 10 |
"total_flos": 42322071132.0,
|
| 11 |
"train_loss": 4.595640664920211,
|
| 12 |
-
"train_runtime":
|
| 13 |
"train_samples": 256,
|
| 14 |
-
"train_samples_per_second": 1.
|
| 15 |
-
"train_steps_per_second": 1.
|
| 16 |
}
|
|
|
|
| 9 |
"temperature_start": 2.5,
|
| 10 |
"total_flos": 42322071132.0,
|
| 11 |
"train_loss": 4.595640664920211,
|
| 12 |
+
"train_runtime": 215.0319,
|
| 13 |
"train_samples": 256,
|
| 14 |
+
"train_samples_per_second": 1.191,
|
| 15 |
+
"train_steps_per_second": 1.191
|
| 16 |
}
|
run_summary.json
CHANGED
|
@@ -10,13 +10,13 @@
|
|
| 10 |
"distill_enabled": true,
|
| 11 |
"enforce_inhouse_models": true,
|
| 12 |
"eval_rows": 64,
|
| 13 |
-
"finished_at": "2026-
|
| 14 |
"fp16": false,
|
| 15 |
"gradient_checkpointing": true,
|
| 16 |
"kd_weight_end": 0.5,
|
| 17 |
"kd_weight_start": 0.65,
|
| 18 |
-
"live_events_path": "/app/runs/
|
| 19 |
-
"live_progress_path": "/app/runs/
|
| 20 |
"lora_alpha": 64,
|
| 21 |
"lora_dropout": 0.05,
|
| 22 |
"lora_enabled": true,
|
|
@@ -34,7 +34,7 @@
|
|
| 34 |
"mps_available": false,
|
| 35 |
"native_strict_mode": false,
|
| 36 |
"native_trainer_mode": true,
|
| 37 |
-
"output_dir": "/app/runs/
|
| 38 |
"requested_warmup_steps": 0,
|
| 39 |
"required_bins": [
|
| 40 |
"bash",
|
|
@@ -51,7 +51,7 @@
|
|
| 51 |
"cuda_total_memory_gb_0": null,
|
| 52 |
"mps_available": false,
|
| 53 |
"torch_available": true,
|
| 54 |
-
"torch_version": "2.
|
| 55 |
},
|
| 56 |
"runtime_system": {
|
| 57 |
"arch": "x86_64",
|
|
@@ -61,13 +61,13 @@
|
|
| 61 |
"git": "/usr/bin/git",
|
| 62 |
"python3": "/usr/local/bin/python3"
|
| 63 |
},
|
| 64 |
-
"collected_at": "2026-
|
| 65 |
"cpu_count": 16,
|
| 66 |
-
"disk_free_gb":
|
| 67 |
"disk_total_gb": 1699.93,
|
| 68 |
"in_venv": false,
|
| 69 |
"is_ubuntu": false,
|
| 70 |
-
"kernel": "6.12.
|
| 71 |
"memory_gb": 123.79,
|
| 72 |
"missing_required_bins": [],
|
| 73 |
"native_mode": true,
|
|
@@ -77,7 +77,7 @@
|
|
| 77 |
"os_id": "debian",
|
| 78 |
"os_name": "Debian GNU/Linux 13 (trixie)",
|
| 79 |
"python_executable": "/usr/local/bin/python",
|
| 80 |
-
"python_version": "3.13.
|
| 81 |
"required_bins": [
|
| 82 |
"bash",
|
| 83 |
"python3",
|
|
@@ -86,7 +86,7 @@
|
|
| 86 |
]
|
| 87 |
},
|
| 88 |
"save_total_limit": 4,
|
| 89 |
-
"system_snapshot_path": "/app/runs/
|
| 90 |
"target_repo_id": "NorthernTribe-Research/UMSR-Reasoner-7B",
|
| 91 |
"teacher_count": 1,
|
| 92 |
"teacher_dtype": "bfloat16",
|
|
|
|
| 10 |
"distill_enabled": true,
|
| 11 |
"enforce_inhouse_models": true,
|
| 12 |
"eval_rows": 64,
|
| 13 |
+
"finished_at": "2026-04-19T07:42:11.481358+00:00",
|
| 14 |
"fp16": false,
|
| 15 |
"gradient_checkpointing": true,
|
| 16 |
"kd_weight_end": 0.5,
|
| 17 |
"kd_weight_start": 0.65,
|
| 18 |
+
"live_events_path": "/app/runs/20260419_073812/live_events.jsonl",
|
| 19 |
+
"live_progress_path": "/app/runs/20260419_073812/live_progress.json",
|
| 20 |
"lora_alpha": 64,
|
| 21 |
"lora_dropout": 0.05,
|
| 22 |
"lora_enabled": true,
|
|
|
|
| 34 |
"mps_available": false,
|
| 35 |
"native_strict_mode": false,
|
| 36 |
"native_trainer_mode": true,
|
| 37 |
+
"output_dir": "/app/runs/20260419_073812",
|
| 38 |
"requested_warmup_steps": 0,
|
| 39 |
"required_bins": [
|
| 40 |
"bash",
|
|
|
|
| 51 |
"cuda_total_memory_gb_0": null,
|
| 52 |
"mps_available": false,
|
| 53 |
"torch_available": true,
|
| 54 |
+
"torch_version": "2.11.0+cu130"
|
| 55 |
},
|
| 56 |
"runtime_system": {
|
| 57 |
"arch": "x86_64",
|
|
|
|
| 61 |
"git": "/usr/bin/git",
|
| 62 |
"python3": "/usr/local/bin/python3"
|
| 63 |
},
|
| 64 |
+
"collected_at": "2026-04-19T07:38:17.456985+00:00",
|
| 65 |
"cpu_count": 16,
|
| 66 |
+
"disk_free_gb": 412.06,
|
| 67 |
"disk_total_gb": 1699.93,
|
| 68 |
"in_venv": false,
|
| 69 |
"is_ubuntu": false,
|
| 70 |
+
"kernel": "6.12.73-95.123.amzn2023.x86_64",
|
| 71 |
"memory_gb": 123.79,
|
| 72 |
"missing_required_bins": [],
|
| 73 |
"native_mode": true,
|
|
|
|
| 77 |
"os_id": "debian",
|
| 78 |
"os_name": "Debian GNU/Linux 13 (trixie)",
|
| 79 |
"python_executable": "/usr/local/bin/python",
|
| 80 |
+
"python_version": "3.13.13",
|
| 81 |
"required_bins": [
|
| 82 |
"bash",
|
| 83 |
"python3",
|
|
|
|
| 86 |
]
|
| 87 |
},
|
| 88 |
"save_total_limit": 4,
|
| 89 |
+
"system_snapshot_path": "/app/runs/20260419_073812/system_snapshot.json",
|
| 90 |
"target_repo_id": "NorthernTribe-Research/UMSR-Reasoner-7B",
|
| 91 |
"teacher_count": 1,
|
| 92 |
"teacher_dtype": "bfloat16",
|
system_snapshot.json
CHANGED
|
@@ -6,13 +6,13 @@
|
|
| 6 |
"git": "/usr/bin/git",
|
| 7 |
"python3": "/usr/local/bin/python3"
|
| 8 |
},
|
| 9 |
-
"collected_at": "2026-
|
| 10 |
"cpu_count": 16,
|
| 11 |
-
"disk_free_gb":
|
| 12 |
"disk_total_gb": 1699.93,
|
| 13 |
"in_venv": false,
|
| 14 |
"is_ubuntu": false,
|
| 15 |
-
"kernel": "6.12.
|
| 16 |
"memory_gb": 123.79,
|
| 17 |
"missing_required_bins": [],
|
| 18 |
"native_mode": true,
|
|
@@ -22,7 +22,7 @@
|
|
| 22 |
"os_id": "debian",
|
| 23 |
"os_name": "Debian GNU/Linux 13 (trixie)",
|
| 24 |
"python_executable": "/usr/local/bin/python",
|
| 25 |
-
"python_version": "3.13.
|
| 26 |
"required_bins": [
|
| 27 |
"bash",
|
| 28 |
"python3",
|
|
|
|
| 6 |
"git": "/usr/bin/git",
|
| 7 |
"python3": "/usr/local/bin/python3"
|
| 8 |
},
|
| 9 |
+
"collected_at": "2026-04-19T07:38:17.456985+00:00",
|
| 10 |
"cpu_count": 16,
|
| 11 |
+
"disk_free_gb": 412.06,
|
| 12 |
"disk_total_gb": 1699.93,
|
| 13 |
"in_venv": false,
|
| 14 |
"is_ubuntu": false,
|
| 15 |
+
"kernel": "6.12.73-95.123.amzn2023.x86_64",
|
| 16 |
"memory_gb": 123.79,
|
| 17 |
"missing_required_bins": [],
|
| 18 |
"native_mode": true,
|
|
|
|
| 22 |
"os_id": "debian",
|
| 23 |
"os_name": "Debian GNU/Linux 13 (trixie)",
|
| 24 |
"python_executable": "/usr/local/bin/python",
|
| 25 |
+
"python_version": "3.13.13",
|
| 26 |
"required_bins": [
|
| 27 |
"bash",
|
| 28 |
"python3",
|
tokenizer_config.json
CHANGED
|
@@ -5,8 +5,12 @@
|
|
| 5 |
"eos_token": "<|endoftext|>",
|
| 6 |
"errors": "replace",
|
| 7 |
"is_local": false,
|
|
|
|
| 8 |
"model_max_length": 1024,
|
| 9 |
"pad_token": "<|endoftext|>",
|
|
|
|
| 10 |
"tokenizer_class": "GPT2Tokenizer",
|
|
|
|
|
|
|
| 11 |
"unk_token": "<|endoftext|>"
|
| 12 |
}
|
|
|
|
| 5 |
"eos_token": "<|endoftext|>",
|
| 6 |
"errors": "replace",
|
| 7 |
"is_local": false,
|
| 8 |
+
"max_length": 512,
|
| 9 |
"model_max_length": 1024,
|
| 10 |
"pad_token": "<|endoftext|>",
|
| 11 |
+
"stride": 0,
|
| 12 |
"tokenizer_class": "GPT2Tokenizer",
|
| 13 |
+
"truncation_side": "right",
|
| 14 |
+
"truncation_strategy": "longest_first",
|
| 15 |
"unk_token": "<|endoftext|>"
|
| 16 |
}
|
trainer_state.json
CHANGED
|
@@ -317,9 +317,9 @@
|
|
| 317 |
"distill_temperature": 2.373046875,
|
| 318 |
"epoch": 0.09765625,
|
| 319 |
"eval_loss": 3.927885055541992,
|
| 320 |
-
"eval_runtime":
|
| 321 |
-
"eval_samples_per_second":
|
| 322 |
-
"eval_steps_per_second":
|
| 323 |
"step": 25
|
| 324 |
},
|
| 325 |
{
|
|
@@ -630,9 +630,9 @@
|
|
| 630 |
"distill_temperature": 2.24609375,
|
| 631 |
"epoch": 0.1953125,
|
| 632 |
"eval_loss": 4.088868141174316,
|
| 633 |
-
"eval_runtime":
|
| 634 |
-
"eval_samples_per_second": 5.
|
| 635 |
-
"eval_steps_per_second": 5.
|
| 636 |
"step": 50
|
| 637 |
},
|
| 638 |
{
|
|
@@ -943,9 +943,9 @@
|
|
| 943 |
"distill_temperature": 2.119140625,
|
| 944 |
"epoch": 0.29296875,
|
| 945 |
"eval_loss": 4.2497992515563965,
|
| 946 |
-
"eval_runtime":
|
| 947 |
-
"eval_samples_per_second":
|
| 948 |
-
"eval_steps_per_second":
|
| 949 |
"step": 75
|
| 950 |
},
|
| 951 |
{
|
|
@@ -1256,9 +1256,9 @@
|
|
| 1256 |
"distill_temperature": 1.9921875,
|
| 1257 |
"epoch": 0.390625,
|
| 1258 |
"eval_loss": 4.411334991455078,
|
| 1259 |
-
"eval_runtime":
|
| 1260 |
-
"eval_samples_per_second": 5.
|
| 1261 |
-
"eval_steps_per_second": 5.
|
| 1262 |
"step": 100
|
| 1263 |
},
|
| 1264 |
{
|
|
@@ -1569,9 +1569,9 @@
|
|
| 1569 |
"distill_temperature": 1.865234375,
|
| 1570 |
"epoch": 0.48828125,
|
| 1571 |
"eval_loss": 4.573975086212158,
|
| 1572 |
-
"eval_runtime":
|
| 1573 |
-
"eval_samples_per_second": 5.
|
| 1574 |
-
"eval_steps_per_second": 5.
|
| 1575 |
"step": 125
|
| 1576 |
},
|
| 1577 |
{
|
|
@@ -1882,9 +1882,9 @@
|
|
| 1882 |
"distill_temperature": 1.73828125,
|
| 1883 |
"epoch": 0.5859375,
|
| 1884 |
"eval_loss": 4.739337921142578,
|
| 1885 |
-
"eval_runtime":
|
| 1886 |
-
"eval_samples_per_second": 5.
|
| 1887 |
-
"eval_steps_per_second": 5.
|
| 1888 |
"step": 150
|
| 1889 |
},
|
| 1890 |
{
|
|
@@ -2195,9 +2195,9 @@
|
|
| 2195 |
"distill_temperature": 1.611328125,
|
| 2196 |
"epoch": 0.68359375,
|
| 2197 |
"eval_loss": 4.90593957901001,
|
| 2198 |
-
"eval_runtime":
|
| 2199 |
-
"eval_samples_per_second":
|
| 2200 |
-
"eval_steps_per_second":
|
| 2201 |
"step": 175
|
| 2202 |
},
|
| 2203 |
{
|
|
@@ -2508,9 +2508,9 @@
|
|
| 2508 |
"distill_temperature": 1.484375,
|
| 2509 |
"epoch": 0.78125,
|
| 2510 |
"eval_loss": 5.072885513305664,
|
| 2511 |
-
"eval_runtime": 12.
|
| 2512 |
-
"eval_samples_per_second": 5.
|
| 2513 |
-
"eval_steps_per_second": 5.
|
| 2514 |
"step": 200
|
| 2515 |
},
|
| 2516 |
{
|
|
@@ -2821,7 +2821,7 @@
|
|
| 2821 |
"distill_temperature": 1.357421875,
|
| 2822 |
"epoch": 0.87890625,
|
| 2823 |
"eval_loss": 5.237745761871338,
|
| 2824 |
-
"eval_runtime": 12.
|
| 2825 |
"eval_samples_per_second": 5.279,
|
| 2826 |
"eval_steps_per_second": 5.279,
|
| 2827 |
"step": 225
|
|
@@ -3134,9 +3134,9 @@
|
|
| 3134 |
"distill_temperature": 1.23046875,
|
| 3135 |
"epoch": 0.9765625,
|
| 3136 |
"eval_loss": 5.399942398071289,
|
| 3137 |
-
"eval_runtime": 12.
|
| 3138 |
-
"eval_samples_per_second": 5.
|
| 3139 |
-
"eval_steps_per_second": 5.
|
| 3140 |
"step": 250
|
| 3141 |
},
|
| 3142 |
{
|
|
@@ -3212,18 +3212,31 @@
|
|
| 3212 |
"step": 256
|
| 3213 |
},
|
| 3214 |
{
|
| 3215 |
-
"distill_ce_loss": 10.
|
| 3216 |
-
"distill_ce_weight": 0.
|
| 3217 |
-
"distill_kd_loss": 0.
|
| 3218 |
-
"distill_kd_weight": 0.
|
| 3219 |
-
"distill_temperature": 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3220 |
"epoch": 1.0,
|
| 3221 |
"step": 256,
|
| 3222 |
"total_flos": 42322071132.0,
|
| 3223 |
"train_loss": 4.595640664920211,
|
| 3224 |
-
"train_runtime":
|
| 3225 |
-
"train_samples_per_second": 1.
|
| 3226 |
-
"train_steps_per_second": 1.
|
| 3227 |
}
|
| 3228 |
],
|
| 3229 |
"logging_steps": 1,
|
|
|
|
| 317 |
"distill_temperature": 2.373046875,
|
| 318 |
"epoch": 0.09765625,
|
| 319 |
"eval_loss": 3.927885055541992,
|
| 320 |
+
"eval_runtime": 12.7363,
|
| 321 |
+
"eval_samples_per_second": 5.025,
|
| 322 |
+
"eval_steps_per_second": 5.025,
|
| 323 |
"step": 25
|
| 324 |
},
|
| 325 |
{
|
|
|
|
| 630 |
"distill_temperature": 2.24609375,
|
| 631 |
"epoch": 0.1953125,
|
| 632 |
"eval_loss": 4.088868141174316,
|
| 633 |
+
"eval_runtime": 10.7991,
|
| 634 |
+
"eval_samples_per_second": 5.926,
|
| 635 |
+
"eval_steps_per_second": 5.926,
|
| 636 |
"step": 50
|
| 637 |
},
|
| 638 |
{
|
|
|
|
| 943 |
"distill_temperature": 2.119140625,
|
| 944 |
"epoch": 0.29296875,
|
| 945 |
"eval_loss": 4.2497992515563965,
|
| 946 |
+
"eval_runtime": 10.459,
|
| 947 |
+
"eval_samples_per_second": 6.119,
|
| 948 |
+
"eval_steps_per_second": 6.119,
|
| 949 |
"step": 75
|
| 950 |
},
|
| 951 |
{
|
|
|
|
| 1256 |
"distill_temperature": 1.9921875,
|
| 1257 |
"epoch": 0.390625,
|
| 1258 |
"eval_loss": 4.411334991455078,
|
| 1259 |
+
"eval_runtime": 10.7332,
|
| 1260 |
+
"eval_samples_per_second": 5.963,
|
| 1261 |
+
"eval_steps_per_second": 5.963,
|
| 1262 |
"step": 100
|
| 1263 |
},
|
| 1264 |
{
|
|
|
|
| 1569 |
"distill_temperature": 1.865234375,
|
| 1570 |
"epoch": 0.48828125,
|
| 1571 |
"eval_loss": 4.573975086212158,
|
| 1572 |
+
"eval_runtime": 11.3539,
|
| 1573 |
+
"eval_samples_per_second": 5.637,
|
| 1574 |
+
"eval_steps_per_second": 5.637,
|
| 1575 |
"step": 125
|
| 1576 |
},
|
| 1577 |
{
|
|
|
|
| 1882 |
"distill_temperature": 1.73828125,
|
| 1883 |
"epoch": 0.5859375,
|
| 1884 |
"eval_loss": 4.739337921142578,
|
| 1885 |
+
"eval_runtime": 11.6684,
|
| 1886 |
+
"eval_samples_per_second": 5.485,
|
| 1887 |
+
"eval_steps_per_second": 5.485,
|
| 1888 |
"step": 150
|
| 1889 |
},
|
| 1890 |
{
|
|
|
|
| 2195 |
"distill_temperature": 1.611328125,
|
| 2196 |
"epoch": 0.68359375,
|
| 2197 |
"eval_loss": 4.90593957901001,
|
| 2198 |
+
"eval_runtime": 11.509,
|
| 2199 |
+
"eval_samples_per_second": 5.561,
|
| 2200 |
+
"eval_steps_per_second": 5.561,
|
| 2201 |
"step": 175
|
| 2202 |
},
|
| 2203 |
{
|
|
|
|
| 2508 |
"distill_temperature": 1.484375,
|
| 2509 |
"epoch": 0.78125,
|
| 2510 |
"eval_loss": 5.072885513305664,
|
| 2511 |
+
"eval_runtime": 12.3769,
|
| 2512 |
+
"eval_samples_per_second": 5.171,
|
| 2513 |
+
"eval_steps_per_second": 5.171,
|
| 2514 |
"step": 200
|
| 2515 |
},
|
| 2516 |
{
|
|
|
|
| 2821 |
"distill_temperature": 1.357421875,
|
| 2822 |
"epoch": 0.87890625,
|
| 2823 |
"eval_loss": 5.237745761871338,
|
| 2824 |
+
"eval_runtime": 12.1232,
|
| 2825 |
"eval_samples_per_second": 5.279,
|
| 2826 |
"eval_steps_per_second": 5.279,
|
| 2827 |
"step": 225
|
|
|
|
| 3134 |
"distill_temperature": 1.23046875,
|
| 3135 |
"epoch": 0.9765625,
|
| 3136 |
"eval_loss": 5.399942398071289,
|
| 3137 |
+
"eval_runtime": 12.2026,
|
| 3138 |
+
"eval_samples_per_second": 5.245,
|
| 3139 |
+
"eval_steps_per_second": 5.245,
|
| 3140 |
"step": 250
|
| 3141 |
},
|
| 3142 |
{
|
|
|
|
| 3212 |
"step": 256
|
| 3213 |
},
|
| 3214 |
{
|
| 3215 |
+
"distill_ce_loss": 10.826531410217285,
|
| 3216 |
+
"distill_ce_weight": 0.5,
|
| 3217 |
+
"distill_kd_loss": 0.0498046875,
|
| 3218 |
+
"distill_kd_weight": 0.5,
|
| 3219 |
+
"distill_temperature": 1.2,
|
| 3220 |
+
"epoch": 1.0,
|
| 3221 |
+
"eval_loss": 5.438441753387451,
|
| 3222 |
+
"eval_runtime": 11.7214,
|
| 3223 |
+
"eval_samples_per_second": 5.46,
|
| 3224 |
+
"eval_steps_per_second": 5.46,
|
| 3225 |
+
"step": 256
|
| 3226 |
+
},
|
| 3227 |
+
{
|
| 3228 |
+
"distill_ce_loss": 10.826531410217285,
|
| 3229 |
+
"distill_ce_weight": 0.5,
|
| 3230 |
+
"distill_kd_loss": 0.0498046875,
|
| 3231 |
+
"distill_kd_weight": 0.5,
|
| 3232 |
+
"distill_temperature": 1.2,
|
| 3233 |
"epoch": 1.0,
|
| 3234 |
"step": 256,
|
| 3235 |
"total_flos": 42322071132.0,
|
| 3236 |
"train_loss": 4.595640664920211,
|
| 3237 |
+
"train_runtime": 215.0319,
|
| 3238 |
+
"train_samples_per_second": 1.191,
|
| 3239 |
+
"train_steps_per_second": 1.191
|
| 3240 |
}
|
| 3241 |
],
|
| 3242 |
"logging_steps": 1,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5201
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8ce0f0a70cf6174b16c45fbe60c590964eea9e019ba10f982d73ee84519b2744
|
| 3 |
size 5201
|