Upup-ashton-wang's group workspace
Hypothesis: Generality
What makes this group special?
Tags
Resa-STILL2Open-S1
Notes
Author
State
Finished
Start time
May 30th, 2025 7:15:09 PM
Runtime
18m 24s
Tracked hours
18m 24s
Run path
upup-ashton-wang-usc/Resa/zfa90l2p
OS
Linux-5.15.0-92-generic-x86_64-with-glibc2.35
Python version
CPython 3.10.16
Command
/home/omer/shangshang/workspace/reasoning/reasoning-sae/./scripts/train/run_sae_based_distill.py --config ./recipes/DeepSeek-R1-Distill-Qwen-1.5B/grpo/distill_curated_still.yaml --host_model_checkpoint checkpoint-2000 --student_model_name DeepSeek-R1-Distill-Qwen-1.5B --distill_dataset_name curated_open_rs1 --distill_type sft_r1_distill --sae_hookpoint model.layers.12 --sae_type finetuned
System Hardware
CPU count | 32 |
Logical CPU count | 64 |
GPU count | 8 |
GPU type | NVIDIA RTX 6000 Ada Generation |
W&B CLI Version
0.19.8
Config
Config parameters are your model's inputs. Learn more
- {} 20 keys▶
- "DeepSeek-R1-Distill-Qwen-1.5B"
- 1
- "curated_open_rs1"
- "sft_r1_distill"
- "checkpoint-2000"
- "curated_still"
- "grpo"
- 0.000001
- 1
- 128
- 0.05
- 32
- [] 7 items▶
- 2
- "model.layers.12"
- "sae-DeepSeek-R1-Distill-Qwen-1.5B-65k"
- "finetuned"
- 500
- 42
- "DeepSeek-R1-Distill-Qwen-1.5B"
Summary
Summary metrics are your model's outputs. Learn more
- {} 5 keys▶
- 1
- 315.7384760112888
- 0.000001
- 16.375
- 296
Artifact Outputs
This run produced these artifacts as outputs. Total: 1. Learn more
Type
Name
Consumer count
Loading...