HansBug/qwen3-8b-mt10-i343

qwen3-8b mt10-mixed-s6 iter 0000343

PyTorch Distributed Checkpoint (DCP) snapshot from an internal RL training run.

Source

• Original path: terminal-rl_qwen3-8b_8gpu_mixed-s6_asb2_ah2-rwnone_dense_dense-c0_dapo-ch0.28-tok1-dyn1_harness-camel-agent_mt10_2026-06-04_124602_0609copy/iter_0000343
• Run started: 2026-06-04 12:46:02
• Iteration: 343

Key training config (parsed from run name)

field	value
base model	qwen3-8b
world size	8 GPU
RL algo	mixed-s6, dapo (ch=0.28, tok=1, dyn=1)
reward	dense, dense-c0
advantage	asb2, ah2-rwnone
harness	camel-agent
variant	mt10

Files

8 sharded .distcp files (~13.35 GiB each) + .metadata + common.pt + metadata.json. Backend: torch_dist (sharded), torch (common). Total ~107 GiB.

__{rank}_{shard}.distcp   for rank in 0..3, shard in 0..1
.metadata
common.pt
metadata.json

Loading

import torch.distributed.checkpoint as dcp
state_dict = {...}  # template with empty tensors of expected shapes
dcp.load(state_dict, checkpoint_id="path/to/iter_0000343")

Or convert to a single torch.save file:

python -m torch.distributed.checkpoint.format_utils \
    dcp_to_torch_save path/to/iter_0000343 consolidated.pt

Companion run: HansBug/qwen3-8b-spear-i375.