abhid1234/qwen-0.5b-tool-agent-grpo
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
2e4a8d6a8318f7393848c4b353077dab400dc0a4
qwen-0.5b-tool-agent-grpo/artifacts/training.log
ModelHub XC 2e4a8d6a83 初始化项目，由ModelHub XC社区提供模型 
Model: abhid1234/qwen-0.5b-tool-agent-grpo
Source: Original Platform
2026-04-28 07:59:12 +08:00

1055 lines
579 KiB
Plaintext
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

Skipping import of cpp extensions due to incompatible torch version 2.10.0+cu128 for torchao version 0.15.0 Please see https://github.com/pytorch/ao/issues/2919 for more info
Loaded 200 train, 50 val scenarios
GRPO config: 4 scenarios/step × 8 rollouts/scenario = 32 rollouts/step
Skipping import of cpp extensions due to incompatible torch version 2.10.0+cu128 for torchao version 0.15.0 Please see https://github.com/pytorch/ao/issues/2919 for more info
/usr/local/lib/python3.12/dist-packages/art/__init__.py:37: UserWarning: WARNING: Unsloth should be imported before [transformers] to ensure all optimizations are applied. Your code may run slower or encounter memory issues without these optimizations.
Please restructure your imports with 'import unsloth' at the top of your file.
import unsloth # noqa: F401
🦥 Unsloth: Will patch your computer to enable 2x faster free finetuning.
🦥 Unsloth Zoo will now patch everything to make training faster!
==((====))== Unsloth 2026.3.3: Fast Qwen2 patching. Transformers: 5.2.0. vLLM: 0.17.0+art1.
\\ /| NVIDIA A100-SXM4-80GB. Num GPUs = 1. Max memory: 79.252 GB. Platform: Linux.
O^O/ \_/ \ Torch: 2.10.0+cu128. CUDA: 8.0. CUDA Toolkit: 12.8. Triton: 3.6.0
\ / Bfloat16 = TRUE. FA [Xformers = 0.0.35. FA2 = False]
"-____-" Free license: http://github.com/unslothai/unsloth
Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!
Loading weights: 0%| | 0/290 [00:00<?, ?it/s]
Loading weights: 0%| | 1/290 [00:00<00:00, 6626.07it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 0%| | 1/290 [00:00<00:00, 3975.64it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/290 [00:00<00:58, 4.92it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/290 [00:00<00:58, 4.92it/s, Materializing param=model.layers.0.input_layernorm.weight]
Loading weights: 1%| | 2/290 [00:00<00:58, 4.92it/s, Materializing param=model.layers.0.input_layernorm.weight]
Loading weights: 1%| | 3/290 [00:00<00:58, 4.92it/s, Materializing param=model.layers.0.mlp.down_proj.weight]
Loading weights: 1%| | 3/290 [00:00<00:58, 4.92it/s, Materializing param=model.layers.0.mlp.down_proj.weight]
Loading weights: 1%|▏ | 4/290 [00:00<00:58, 4.92it/s, Materializing param=model.layers.0.mlp.gate_proj.weight]
Loading weights: 1%|▏ | 4/290 [00:00<00:58, 4.92it/s, Materializing param=model.layers.0.mlp.gate_proj.weight]
Loading weights: 2%|▏ | 5/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.mlp.up_proj.weight]
Loading weights: 2%|▏ | 5/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.mlp.up_proj.weight]
Loading weights: 2%|▏ | 6/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.post_attention_layernorm.weight]
Loading weights: 2%|▏ | 6/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.post_attention_layernorm.weight]
Loading weights: 2%|▏ | 7/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.self_attn.k_proj.bias]
Loading weights: 2%|▏ | 7/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.self_attn.k_proj.bias]
Loading weights: 3%|▎ | 8/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.self_attn.k_proj.weight]
Loading weights: 3%|▎ | 8/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.self_attn.k_proj.weight]
Loading weights: 3%|▎ | 9/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.self_attn.o_proj.weight]
Loading weights: 3%|▎ | 9/290 [00:00<00:57, 4.92it/s, Materializing param=model.layers.0.self_attn.o_proj.weight]
Loading weights: 3%|▎ | 10/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.0.self_attn.q_proj.bias]
Loading weights: 3%|▎ | 10/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.0.self_attn.q_proj.bias]
Loading weights: 4%|▍ | 11/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.0.self_attn.q_proj.weight]
Loading weights: 4%|▍ | 11/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.0.self_attn.q_proj.weight]
Loading weights: 4%|▍ | 12/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.0.self_attn.v_proj.bias]
Loading weights: 4%|▍ | 12/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.0.self_attn.v_proj.bias]
Loading weights: 4%|▍ | 13/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.0.self_attn.v_proj.weight]
Loading weights: 4%|▍ | 13/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.0.self_attn.v_proj.weight]
Loading weights: 5%|▍ | 14/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.1.input_layernorm.weight]
Loading weights: 5%|▍ | 14/290 [00:00<00:56, 4.92it/s, Materializing param=model.layers.1.input_layernorm.weight]
Loading weights: 5%|▌ | 15/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.mlp.down_proj.weight]
Loading weights: 5%|▌ | 15/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.mlp.down_proj.weight]
Loading weights: 6%|▌ | 16/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.mlp.gate_proj.weight]
Loading weights: 6%|▌ | 16/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.mlp.gate_proj.weight]
Loading weights: 6%|▌ | 17/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.mlp.up_proj.weight]
Loading weights: 6%|▌ | 17/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.mlp.up_proj.weight]
Loading weights: 6%|▌ | 18/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.post_attention_layernorm.weight]
Loading weights: 6%|▌ | 18/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.post_attention_layernorm.weight]
Loading weights: 7%|▋ | 19/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.self_attn.k_proj.bias]
Loading weights: 7%|▋ | 19/290 [00:00<00:55, 4.92it/s, Materializing param=model.layers.1.self_attn.k_proj.bias]
Loading weights: 7%|▋ | 20/290 [00:00<00:54, 4.92it/s, Materializing param=model.layers.1.self_attn.k_proj.weight]
Loading weights: 7%|▋ | 20/290 [00:00<00:54, 4.92it/s, Materializing param=model.layers.1.self_attn.k_proj.weight]
Loading weights: 7%|▋ | 21/290 [00:00<00:54, 4.92it/s, Materializing param=model.layers.1.self_attn.o_proj.weight]
Loading weights: 7%|▋ | 21/290 [00:00<00:54, 4.92it/s, Materializing param=model.layers.1.self_attn.o_proj.weight]
Loading weights: 8%|▊ | 22/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.o_proj.weight]
Loading weights: 8%|▊ | 22/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.q_proj.bias]
Loading weights: 8%|▊ | 22/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.q_proj.bias]
Loading weights: 8%|▊ | 23/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.q_proj.weight]
Loading weights: 8%|▊ | 23/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.q_proj.weight]
Loading weights: 8%|▊ | 24/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.v_proj.bias]
Loading weights: 8%|▊ | 24/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.v_proj.bias]
Loading weights: 9%|▊ | 25/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.v_proj.weight]
Loading weights: 9%|▊ | 25/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.1.self_attn.v_proj.weight]
Loading weights: 9%|▉ | 26/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.input_layernorm.weight]
Loading weights: 9%|▉ | 26/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.input_layernorm.weight]
Loading weights: 9%|▉ | 27/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.mlp.down_proj.weight]
Loading weights: 9%|▉ | 27/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.mlp.down_proj.weight]
Loading weights: 10%|▉ | 28/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.mlp.gate_proj.weight]
Loading weights: 10%|▉ | 28/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.mlp.gate_proj.weight]
Loading weights: 10%|█ | 29/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.mlp.up_proj.weight]
Loading weights: 10%|█ | 29/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.mlp.up_proj.weight]
Loading weights: 10%|█ | 30/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.post_attention_layernorm.weight]
Loading weights: 10%|█ | 30/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.post_attention_layernorm.weight]
Loading weights: 11%|█ | 31/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.k_proj.bias]
Loading weights: 11%|█ | 31/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.k_proj.bias]
Loading weights: 11%|█ | 32/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.k_proj.weight]
Loading weights: 11%|█ | 32/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.k_proj.weight]
Loading weights: 11%|█▏ | 33/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.o_proj.weight]
Loading weights: 11%|█▏ | 33/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.o_proj.weight]
Loading weights: 12%|█▏ | 34/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.q_proj.bias]
Loading weights: 12%|█▏ | 34/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.q_proj.bias]
Loading weights: 12%|█▏ | 35/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.q_proj.weight]
Loading weights: 12%|█▏ | 35/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.q_proj.weight]
Loading weights: 12%|█▏ | 36/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.v_proj.bias]
Loading weights: 12%|█▏ | 36/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.v_proj.bias]
Loading weights: 13%|█▎ | 37/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.v_proj.weight]
Loading weights: 13%|█▎ | 37/290 [00:00<00:04, 55.55it/s, Materializing param=model.layers.2.self_attn.v_proj.weight]
Loading weights: 13%|█▎ | 38/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.2.self_attn.v_proj.weight]
Loading weights: 13%|█▎ | 38/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.input_layernorm.weight]
Loading weights: 13%|█▎ | 38/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.input_layernorm.weight]
Loading weights: 13%|█▎ | 39/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.mlp.down_proj.weight]
Loading weights: 13%|█▎ | 39/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.mlp.down_proj.weight]
Loading weights: 14%|█▍ | 40/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.mlp.gate_proj.weight]
Loading weights: 14%|█▍ | 40/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.mlp.gate_proj.weight]
Loading weights: 14%|█▍ | 41/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.mlp.up_proj.weight]
Loading weights: 14%|█▍ | 41/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.mlp.up_proj.weight]
Loading weights: 14%|█▍ | 42/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.post_attention_layernorm.weight]
Loading weights: 14%|█▍ | 42/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.post_attention_layernorm.weight]
Loading weights: 15%|█▍ | 43/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.k_proj.bias]
Loading weights: 15%|█▍ | 43/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.k_proj.bias]
Loading weights: 15%|█▌ | 44/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.k_proj.weight]
Loading weights: 15%|█▌ | 44/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.k_proj.weight]
Loading weights: 16%|█▌ | 45/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.o_proj.weight]
Loading weights: 16%|█▌ | 45/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.o_proj.weight]
Loading weights: 16%|█▌ | 46/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.q_proj.bias]
Loading weights: 16%|█▌ | 46/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.q_proj.bias]
Loading weights: 16%|█▌ | 47/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.q_proj.weight]
Loading weights: 16%|█▌ | 47/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.q_proj.weight]
Loading weights: 17%|█▋ | 48/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.v_proj.bias]
Loading weights: 17%|█▋ | 48/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.v_proj.bias]
Loading weights: 17%|█▋ | 49/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.v_proj.weight]
Loading weights: 17%|█▋ | 49/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.3.self_attn.v_proj.weight]
Loading weights: 17%|█▋ | 50/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.input_layernorm.weight]
Loading weights: 17%|█▋ | 50/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.input_layernorm.weight]
Loading weights: 18%|█▊ | 51/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.mlp.down_proj.weight]
Loading weights: 18%|█▊ | 51/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.mlp.down_proj.weight]
Loading weights: 18%|█▊ | 52/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.mlp.gate_proj.weight]
Loading weights: 18%|█▊ | 52/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.mlp.gate_proj.weight]
Loading weights: 18%|█▊ | 53/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.mlp.up_proj.weight]
Loading weights: 18%|█▊ | 53/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.mlp.up_proj.weight]
Loading weights: 19%|█▊ | 54/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.post_attention_layernorm.weight]
Loading weights: 19%|█▊ | 54/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.post_attention_layernorm.weight]
Loading weights: 19%|█▉ | 55/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.k_proj.bias]
Loading weights: 19%|█▉ | 55/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.k_proj.bias]
Loading weights: 19%|█▉ | 56/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.k_proj.weight]
Loading weights: 19%|█▉ | 56/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.k_proj.weight]
Loading weights: 20%|█▉ | 57/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.o_proj.weight]
Loading weights: 20%|█▉ | 57/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.o_proj.weight]
Loading weights: 20%|██ | 58/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.q_proj.bias]
Loading weights: 20%|██ | 58/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.q_proj.bias]
Loading weights: 20%|██ | 59/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.q_proj.weight]
Loading weights: 20%|██ | 59/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.q_proj.weight]
Loading weights: 21%|██ | 60/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.v_proj.bias]
Loading weights: 21%|██ | 60/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.v_proj.bias]
Loading weights: 21%|██ | 61/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.v_proj.weight]
Loading weights: 21%|██ | 61/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.4.self_attn.v_proj.weight]
Loading weights: 21%|██▏ | 62/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.input_layernorm.weight]
Loading weights: 21%|██▏ | 62/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.input_layernorm.weight]
Loading weights: 22%|██▏ | 63/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.mlp.down_proj.weight]
Loading weights: 22%|██▏ | 63/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.mlp.down_proj.weight]
Loading weights: 22%|██▏ | 64/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.mlp.gate_proj.weight]
Loading weights: 22%|██▏ | 64/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.mlp.gate_proj.weight]
Loading weights: 22%|██▏ | 65/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.mlp.up_proj.weight]
Loading weights: 22%|██▏ | 65/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.mlp.up_proj.weight]
Loading weights: 23%|██▎ | 66/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.post_attention_layernorm.weight]
Loading weights: 23%|██▎ | 66/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.post_attention_layernorm.weight]
Loading weights: 23%|██▎ | 67/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.k_proj.bias]
Loading weights: 23%|██▎ | 67/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.k_proj.bias]
Loading weights: 23%|██▎ | 68/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.k_proj.weight]
Loading weights: 23%|██▎ | 68/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.k_proj.weight]
Loading weights: 24%|██▍ | 69/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.o_proj.weight]
Loading weights: 24%|██▍ | 69/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.o_proj.weight]
Loading weights: 24%|██▍ | 70/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.q_proj.bias]
Loading weights: 24%|██▍ | 70/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.q_proj.bias]
Loading weights: 24%|██▍ | 71/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.q_proj.weight]
Loading weights: 24%|██▍ | 71/290 [00:00<00:03, 72.81it/s, Materializing param=model.layers.5.self_attn.q_proj.weight]
Loading weights: 25%|██▍ | 72/290 [00:00<00:02, 72.81it/s, Materializing param=model.layers.5.self_attn.v_proj.bias]
Loading weights: 25%|██▍ | 72/290 [00:00<00:02, 72.81it/s, Materializing param=model.layers.5.self_attn.v_proj.bias]
Loading weights: 25%|██▌ | 73/290 [00:00<00:02, 72.81it/s, Materializing param=model.layers.5.self_attn.v_proj.weight]
Loading weights: 25%|██▌ | 73/290 [00:00<00:02, 72.81it/s, Materializing param=model.layers.5.self_attn.v_proj.weight]
Loading weights: 26%|██▌ | 74/290 [00:00<00:02, 72.81it/s, Materializing param=model.layers.6.input_layernorm.weight]
Loading weights: 26%|██▌ | 74/290 [00:00<00:02, 72.81it/s, Materializing param=model.layers.6.input_layernorm.weight]
Loading weights: 26%|██▌ | 75/290 [00:00<00:02, 72.81it/s, Materializing param=model.layers.6.mlp.down_proj.weight]
Loading weights: 26%|██▌ | 75/290 [00:00<00:02, 72.81it/s, Materializing param=model.layers.6.mlp.down_proj.weight]
Loading weights: 26%|██▌ | 76/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.mlp.down_proj.weight]
Loading weights: 26%|██▌ | 76/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.mlp.gate_proj.weight]
Loading weights: 26%|██▌ | 76/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.mlp.gate_proj.weight]
Loading weights: 27%|██▋ | 77/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.mlp.up_proj.weight]
Loading weights: 27%|██▋ | 77/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.mlp.up_proj.weight]
Loading weights: 27%|██▋ | 78/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.post_attention_layernorm.weight]
Loading weights: 27%|██▋ | 78/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.post_attention_layernorm.weight]
Loading weights: 27%|██▋ | 79/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.k_proj.bias]
Loading weights: 27%|██▋ | 79/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.k_proj.bias]
Loading weights: 28%|██▊ | 80/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.k_proj.weight]
Loading weights: 28%|██▊ | 80/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.k_proj.weight]
Loading weights: 28%|██▊ | 81/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.o_proj.weight]
Loading weights: 28%|██▊ | 81/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.o_proj.weight]
Loading weights: 28%|██▊ | 82/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.q_proj.bias]
Loading weights: 28%|██▊ | 82/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.q_proj.bias]
Loading weights: 29%|██▊ | 83/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.q_proj.weight]
Loading weights: 29%|██▊ | 83/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.q_proj.weight]
Loading weights: 29%|██▉ | 84/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.v_proj.bias]
Loading weights: 29%|██▉ | 84/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.v_proj.bias]
Loading weights: 29%|██▉ | 85/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.v_proj.weight]
Loading weights: 29%|██▉ | 85/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.6.self_attn.v_proj.weight]
Loading weights: 30%|██▉ | 86/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.input_layernorm.weight]
Loading weights: 30%|██▉ | 86/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.input_layernorm.weight]
Loading weights: 30%|███ | 87/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.mlp.down_proj.weight]
Loading weights: 30%|███ | 87/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.mlp.down_proj.weight]
Loading weights: 30%|███ | 88/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.mlp.gate_proj.weight]
Loading weights: 30%|███ | 88/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.mlp.gate_proj.weight]
Loading weights: 31%|███ | 89/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.mlp.up_proj.weight]
Loading weights: 31%|███ | 89/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.mlp.up_proj.weight]
Loading weights: 31%|███ | 90/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.post_attention_layernorm.weight]
Loading weights: 31%|███ | 90/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.post_attention_layernorm.weight]
Loading weights: 31%|███▏ | 91/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.k_proj.bias]
Loading weights: 31%|███▏ | 91/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.k_proj.bias]
Loading weights: 32%|███▏ | 92/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.k_proj.weight]
Loading weights: 32%|███▏ | 92/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.k_proj.weight]
Loading weights: 32%|███▏ | 93/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.o_proj.weight]
Loading weights: 32%|███▏ | 93/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.o_proj.weight]
Loading weights: 32%|███▏ | 94/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.q_proj.bias]
Loading weights: 32%|███▏ | 94/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.q_proj.bias]
Loading weights: 33%|███▎ | 95/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.q_proj.weight]
Loading weights: 33%|███▎ | 95/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.q_proj.weight]
Loading weights: 33%|███▎ | 96/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.v_proj.bias]
Loading weights: 33%|███▎ | 96/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.v_proj.bias]
Loading weights: 33%|███▎ | 97/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.v_proj.weight]
Loading weights: 33%|███▎ | 97/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.7.self_attn.v_proj.weight]
Loading weights: 34%|███▍ | 98/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.input_layernorm.weight]
Loading weights: 34%|███▍ | 98/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.input_layernorm.weight]
Loading weights: 34%|███▍ | 99/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.mlp.down_proj.weight]
Loading weights: 34%|███▍ | 99/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.mlp.down_proj.weight]
Loading weights: 34%|███▍ | 100/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.mlp.gate_proj.weight]
Loading weights: 34%|███▍ | 100/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.mlp.gate_proj.weight]
Loading weights: 35%|███▍ | 101/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.mlp.up_proj.weight]
Loading weights: 35%|███▍ | 101/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.mlp.up_proj.weight]
Loading weights: 35%|███▌ | 102/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.post_attention_layernorm.weight]
Loading weights: 35%|███▌ | 102/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.post_attention_layernorm.weight]
Loading weights: 36%|███▌ | 103/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.k_proj.bias]
Loading weights: 36%|███▌ | 103/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.k_proj.bias]
Loading weights: 36%|███▌ | 104/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.k_proj.weight]
Loading weights: 36%|███▌ | 104/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.k_proj.weight]
Loading weights: 36%|███▌ | 105/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.o_proj.weight]
Loading weights: 36%|███▌ | 105/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.o_proj.weight]
Loading weights: 37%|███▋ | 106/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.q_proj.bias]
Loading weights: 37%|███▋ | 106/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.q_proj.bias]
Loading weights: 37%|███▋ | 107/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.q_proj.weight]
Loading weights: 37%|███▋ | 107/290 [00:00<00:01, 149.12it/s, Materializing param=model.layers.8.self_attn.q_proj.weight]
Loading weights: 37%|███▋ | 108/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.8.self_attn.q_proj.weight]
Loading weights: 37%|███▋ | 108/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.8.self_attn.v_proj.bias]
Loading weights: 37%|███▋ | 108/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.8.self_attn.v_proj.bias]
Loading weights: 38%|███▊ | 109/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.8.self_attn.v_proj.weight]
Loading weights: 38%|███▊ | 109/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.8.self_attn.v_proj.weight]
Loading weights: 38%|███▊ | 110/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.input_layernorm.weight]
Loading weights: 38%|███▊ | 110/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.input_layernorm.weight]
Loading weights: 38%|███▊ | 111/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.mlp.down_proj.weight]
Loading weights: 38%|███▊ | 111/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.mlp.down_proj.weight]
Loading weights: 39%|███▊ | 112/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.mlp.gate_proj.weight]
Loading weights: 39%|███▊ | 112/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.mlp.gate_proj.weight]
Loading weights: 39%|███▉ | 113/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.mlp.up_proj.weight]
Loading weights: 39%|███▉ | 113/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.mlp.up_proj.weight]
Loading weights: 39%|███▉ | 114/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.post_attention_layernorm.weight]
Loading weights: 39%|███▉ | 114/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.post_attention_layernorm.weight]
Loading weights: 40%|███▉ | 115/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.k_proj.bias]
Loading weights: 40%|███▉ | 115/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.k_proj.bias]
Loading weights: 40%|████ | 116/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.k_proj.weight]
Loading weights: 40%|████ | 116/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.k_proj.weight]
Loading weights: 40%|████ | 117/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.o_proj.weight]
Loading weights: 40%|████ | 117/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.o_proj.weight]
Loading weights: 41%|████ | 118/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.q_proj.bias]
Loading weights: 41%|████ | 118/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.q_proj.bias]
Loading weights: 41%|████ | 119/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 41%|████ | 119/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 41%|████▏ | 120/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.v_proj.bias]
Loading weights: 41%|████▏ | 120/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.v_proj.bias]
Loading weights: 42%|████▏ | 121/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.v_proj.weight]
Loading weights: 42%|████▏ | 121/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.9.self_attn.v_proj.weight]
Loading weights: 42%|████▏ | 122/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.input_layernorm.weight]
Loading weights: 42%|████▏ | 122/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.input_layernorm.weight]
Loading weights: 42%|████▏ | 123/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.mlp.down_proj.weight]
Loading weights: 42%|████▏ | 123/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.mlp.down_proj.weight]
Loading weights: 43%|████▎ | 124/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.mlp.gate_proj.weight]
Loading weights: 43%|████▎ | 124/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.mlp.gate_proj.weight]
Loading weights: 43%|████▎ | 125/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 43%|████▎ | 125/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 43%|████▎ | 126/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 43%|████▎ | 126/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 44%|████▍ | 127/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.k_proj.bias]
Loading weights: 44%|████▍ | 127/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.k_proj.bias]
Loading weights: 44%|████▍ | 128/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.k_proj.weight]
Loading weights: 44%|████▍ | 128/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.k_proj.weight]
Loading weights: 44%|████▍ | 129/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.o_proj.weight]
Loading weights: 44%|████▍ | 129/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.o_proj.weight]
Loading weights: 45%|████▍ | 130/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.q_proj.bias]
Loading weights: 45%|████▍ | 130/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.q_proj.bias]
Loading weights: 45%|████▌ | 131/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.q_proj.weight]
Loading weights: 45%|████▌ | 131/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.q_proj.weight]
Loading weights: 46%|████▌ | 132/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.v_proj.bias]
Loading weights: 46%|████▌ | 132/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.v_proj.bias]
Loading weights: 46%|████▌ | 133/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.v_proj.weight]
Loading weights: 46%|████▌ | 133/290 [00:00<00:00, 188.88it/s, Materializing param=model.layers.10.self_attn.v_proj.weight]
Loading weights: 46%|████▌ | 134/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.10.self_attn.v_proj.weight]
Loading weights: 46%|████▌ | 134/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.input_layernorm.weight]
Loading weights: 46%|████▌ | 134/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.input_layernorm.weight]
Loading weights: 47%|████▋ | 135/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.mlp.down_proj.weight]
Loading weights: 47%|████▋ | 135/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.mlp.down_proj.weight]
Loading weights: 47%|████▋ | 136/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.mlp.gate_proj.weight]
Loading weights: 47%|████▋ | 136/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.mlp.gate_proj.weight]
Loading weights: 47%|████▋ | 137/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.mlp.up_proj.weight]
Loading weights: 47%|████▋ | 137/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.mlp.up_proj.weight]
Loading weights: 48%|████▊ | 138/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.post_attention_layernorm.weight]
Loading weights: 48%|████▊ | 138/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.post_attention_layernorm.weight]
Loading weights: 48%|████▊ | 139/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.k_proj.bias]
Loading weights: 48%|████▊ | 139/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.k_proj.bias]
Loading weights: 48%|████▊ | 140/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.k_proj.weight]
Loading weights: 48%|████▊ | 140/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.k_proj.weight]
Loading weights: 49%|████▊ | 141/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.o_proj.weight]
Loading weights: 49%|████▊ | 141/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.o_proj.weight]
Loading weights: 49%|████▉ | 142/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.q_proj.bias]
Loading weights: 49%|████▉ | 142/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.q_proj.bias]
Loading weights: 49%|████▉ | 143/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.q_proj.weight]
Loading weights: 49%|████▉ | 143/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.q_proj.weight]
Loading weights: 50%|████▉ | 144/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.v_proj.bias]
Loading weights: 50%|████▉ | 144/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.v_proj.bias]
Loading weights: 50%|█████ | 145/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 145/290 [00:00<00:00, 206.86it/s, Materializing param=model.layers.11.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 146/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.input_layernorm.weight]
Loading weights: 50%|█████ | 146/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.input_layernorm.weight]
Loading weights: 51%|█████ | 147/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.mlp.down_proj.weight]
Loading weights: 51%|█████ | 147/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.mlp.down_proj.weight]
Loading weights: 51%|█████ | 148/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.mlp.gate_proj.weight]
Loading weights: 51%|█████ | 148/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.mlp.gate_proj.weight]
Loading weights: 51%|█████▏ | 149/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 149/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.mlp.up_proj.weight]
Loading weights: 52%|█████▏ | 150/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 150/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 151/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.k_proj.bias]
Loading weights: 52%|█████▏ | 151/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.k_proj.bias]
Loading weights: 52%|█████▏ | 152/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.k_proj.weight]
Loading weights: 52%|█████▏ | 152/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.k_proj.weight]
Loading weights: 53%|█████▎ | 153/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.o_proj.weight]
Loading weights: 53%|█████▎ | 153/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.o_proj.weight]
Loading weights: 53%|█████▎ | 154/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.q_proj.bias]
Loading weights: 53%|█████▎ | 154/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.q_proj.bias]
Loading weights: 53%|█████▎ | 155/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.q_proj.weight]
Loading weights: 53%|█████▎ | 155/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.q_proj.weight]
Loading weights: 54%|█████▍ | 156/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.v_proj.bias]
Loading weights: 54%|█████▍ | 156/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.v_proj.bias]
Loading weights: 54%|█████▍ | 157/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.v_proj.weight]
Loading weights: 54%|█████▍ | 157/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.12.self_attn.v_proj.weight]
Loading weights: 54%|█████▍ | 158/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.input_layernorm.weight]
Loading weights: 54%|█████▍ | 158/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.input_layernorm.weight]
Loading weights: 55%|█████▍ | 159/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.mlp.down_proj.weight]
Loading weights: 55%|█████▍ | 159/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.mlp.down_proj.weight]
Loading weights: 55%|█████▌ | 160/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.mlp.gate_proj.weight]
Loading weights: 55%|█████▌ | 160/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.mlp.gate_proj.weight]
Loading weights: 56%|█████▌ | 161/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.mlp.up_proj.weight]
Loading weights: 56%|█████▌ | 161/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.mlp.up_proj.weight]
Loading weights: 56%|█████▌ | 162/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 162/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 163/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.k_proj.bias]
Loading weights: 56%|█████▌ | 163/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.k_proj.bias]
Loading weights: 57%|█████▋ | 164/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.k_proj.weight]
Loading weights: 57%|█████▋ | 164/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.k_proj.weight]
Loading weights: 57%|█████▋ | 165/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.o_proj.weight]
Loading weights: 57%|█████▋ | 165/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.o_proj.weight]
Loading weights: 57%|█████▋ | 166/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.q_proj.bias]
Loading weights: 57%|█████▋ | 166/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.q_proj.bias]
Loading weights: 58%|█████▊ | 167/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 58%|█████▊ | 167/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 58%|█████▊ | 168/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.v_proj.bias]
Loading weights: 58%|█████▊ | 168/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.v_proj.bias]
Loading weights: 58%|█████▊ | 169/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 58%|█████▊ | 169/290 [00:01<00:00, 206.86it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 59%|█████▊ | 170/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 59%|█████▊ | 170/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 59%|█████▊ | 170/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 59%|█████▉ | 171/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 59%|█████▉ | 171/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 59%|█████▉ | 172/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.mlp.gate_proj.weight]
Loading weights: 59%|█████▉ | 172/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.mlp.gate_proj.weight]
Loading weights: 60%|█████▉ | 173/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 60%|█████▉ | 173/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 60%|██████ | 174/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 60%|██████ | 174/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 60%|██████ | 175/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.k_proj.bias]
Loading weights: 60%|██████ | 175/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.k_proj.bias]
Loading weights: 61%|██████ | 176/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.k_proj.weight]
Loading weights: 61%|██████ | 176/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.k_proj.weight]
Loading weights: 61%|██████ | 177/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.o_proj.weight]
Loading weights: 61%|██████ | 177/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.o_proj.weight]
Loading weights: 61%|██████▏ | 178/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.q_proj.bias]
Loading weights: 61%|██████▏ | 178/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.q_proj.bias]
Loading weights: 62%|██████▏ | 179/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.q_proj.weight]
Loading weights: 62%|██████▏ | 179/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.q_proj.weight]
Loading weights: 62%|██████▏ | 180/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.v_proj.bias]
Loading weights: 62%|██████▏ | 180/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.v_proj.bias]
Loading weights: 62%|██████▏ | 181/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.v_proj.weight]
Loading weights: 62%|██████▏ | 181/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.14.self_attn.v_proj.weight]
Loading weights: 63%|██████▎ | 182/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.input_layernorm.weight]
Loading weights: 63%|██████▎ | 182/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.input_layernorm.weight]
Loading weights: 63%|██████▎ | 183/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.mlp.down_proj.weight]
Loading weights: 63%|██████▎ | 183/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.mlp.down_proj.weight]
Loading weights: 63%|██████▎ | 184/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.mlp.gate_proj.weight]
Loading weights: 63%|██████▎ | 184/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.mlp.gate_proj.weight]
Loading weights: 64%|██████▍ | 185/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.mlp.up_proj.weight]
Loading weights: 64%|██████▍ | 185/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.mlp.up_proj.weight]
Loading weights: 64%|██████▍ | 186/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 64%|██████▍ | 186/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 64%|██████▍ | 187/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.k_proj.bias]
Loading weights: 64%|██████▍ | 187/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.k_proj.bias]
Loading weights: 65%|██████▍ | 188/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 65%|██████▍ | 188/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 65%|██████▌ | 189/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 65%|██████▌ | 189/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 66%|██████▌ | 190/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.q_proj.bias]
Loading weights: 66%|██████▌ | 190/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.q_proj.bias]
Loading weights: 66%|██████▌ | 191/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.q_proj.weight]
Loading weights: 66%|██████▌ | 191/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.q_proj.weight]
Loading weights: 66%|██████▌ | 192/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.v_proj.bias]
Loading weights: 66%|██████▌ | 192/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.v_proj.bias]
Loading weights: 67%|██████▋ | 193/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.v_proj.weight]
Loading weights: 67%|██████▋ | 193/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.15.self_attn.v_proj.weight]
Loading weights: 67%|██████▋ | 194/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.input_layernorm.weight]
Loading weights: 67%|██████▋ | 194/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.input_layernorm.weight]
Loading weights: 67%|██████▋ | 195/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.mlp.down_proj.weight]
Loading weights: 67%|██████▋ | 195/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.mlp.down_proj.weight]
Loading weights: 68%|██████▊ | 196/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.mlp.gate_proj.weight]
Loading weights: 68%|██████▊ | 196/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.mlp.gate_proj.weight]
Loading weights: 68%|██████▊ | 197/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.mlp.up_proj.weight]
Loading weights: 68%|██████▊ | 197/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.mlp.up_proj.weight]
Loading weights: 68%|██████▊ | 198/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.post_attention_layernorm.weight]
Loading weights: 68%|██████▊ | 198/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.post_attention_layernorm.weight]
Loading weights: 69%|██████▊ | 199/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.self_attn.k_proj.bias]
Loading weights: 69%|██████▊ | 199/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.self_attn.k_proj.bias]
Loading weights: 69%|██████▉ | 200/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.self_attn.k_proj.weight]
Loading weights: 69%|██████▉ | 200/290 [00:01<00:00, 249.02it/s, Materializing param=model.layers.16.self_attn.k_proj.weight]
Loading weights: 69%|██████▉ | 201/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.k_proj.weight]
Loading weights: 69%|██████▉ | 201/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.o_proj.weight]
Loading weights: 69%|██████▉ | 201/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.o_proj.weight]
Loading weights: 70%|██████▉ | 202/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.q_proj.bias]
Loading weights: 70%|██████▉ | 202/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.q_proj.bias]
Loading weights: 70%|███████ | 203/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.q_proj.weight]
Loading weights: 70%|███████ | 203/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.q_proj.weight]
Loading weights: 70%|███████ | 204/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.v_proj.bias]
Loading weights: 70%|███████ | 204/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.v_proj.bias]
Loading weights: 71%|███████ | 205/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.v_proj.weight]
Loading weights: 71%|███████ | 205/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.16.self_attn.v_proj.weight]
Loading weights: 71%|███████ | 206/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.input_layernorm.weight]
Loading weights: 71%|███████ | 206/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.input_layernorm.weight]
Loading weights: 71%|███████▏ | 207/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.mlp.down_proj.weight]
Loading weights: 71%|███████▏ | 207/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.mlp.down_proj.weight]
Loading weights: 72%|███████▏ | 208/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.mlp.gate_proj.weight]
Loading weights: 72%|███████▏ | 208/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.mlp.gate_proj.weight]
Loading weights: 72%|███████▏ | 209/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 72%|███████▏ | 209/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 72%|███████▏ | 210/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 72%|███████▏ | 210/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 73%|███████▎ | 211/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.k_proj.bias]
Loading weights: 73%|███████▎ | 211/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.k_proj.bias]
Loading weights: 73%|███████▎ | 212/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.k_proj.weight]
Loading weights: 73%|███████▎ | 212/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.k_proj.weight]
Loading weights: 73%|███████▎ | 213/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.o_proj.weight]
Loading weights: 73%|███████▎ | 213/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.o_proj.weight]
Loading weights: 74%|███████▍ | 214/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.q_proj.bias]
Loading weights: 74%|███████▍ | 214/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.q_proj.bias]
Loading weights: 74%|███████▍ | 215/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.q_proj.weight]
Loading weights: 74%|███████▍ | 215/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.q_proj.weight]
Loading weights: 74%|███████▍ | 216/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.v_proj.bias]
Loading weights: 74%|███████▍ | 216/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.v_proj.bias]
Loading weights: 75%|███████▍ | 217/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.v_proj.weight]
Loading weights: 75%|███████▍ | 217/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.17.self_attn.v_proj.weight]
Loading weights: 75%|███████▌ | 218/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.input_layernorm.weight]
Loading weights: 75%|███████▌ | 218/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.input_layernorm.weight]
Loading weights: 76%|███████▌ | 219/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.mlp.down_proj.weight]
Loading weights: 76%|███████▌ | 219/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.mlp.down_proj.weight]
Loading weights: 76%|███████▌ | 220/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.mlp.gate_proj.weight]
Loading weights: 76%|███████▌ | 220/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.mlp.gate_proj.weight]
Loading weights: 76%|███████▌ | 221/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 221/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.mlp.up_proj.weight]
Loading weights: 77%|███████▋ | 222/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 222/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 223/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.k_proj.bias]
Loading weights: 77%|███████▋ | 223/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.k_proj.bias]
Loading weights: 77%|███████▋ | 224/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.k_proj.weight]
Loading weights: 77%|███████▋ | 224/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.k_proj.weight]
Loading weights: 78%|███████▊ | 225/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.o_proj.weight]
Loading weights: 78%|███████▊ | 225/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.o_proj.weight]
Loading weights: 78%|███████▊ | 226/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.q_proj.bias]
Loading weights: 78%|███████▊ | 226/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.q_proj.bias]
Loading weights: 78%|███████▊ | 227/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.q_proj.weight]
Loading weights: 78%|███████▊ | 227/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.q_proj.weight]
Loading weights: 79%|███████▊ | 228/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.v_proj.bias]
Loading weights: 79%|███████▊ | 228/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.v_proj.bias]
Loading weights: 79%|███████▉ | 229/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.v_proj.weight]
Loading weights: 79%|███████▉ | 229/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.18.self_attn.v_proj.weight]
Loading weights: 79%|███████▉ | 230/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.input_layernorm.weight]
Loading weights: 79%|███████▉ | 230/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.input_layernorm.weight]
Loading weights: 80%|███████▉ | 231/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.mlp.down_proj.weight]
Loading weights: 80%|███████▉ | 231/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.mlp.down_proj.weight]
Loading weights: 80%|████████ | 232/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.mlp.gate_proj.weight]
Loading weights: 80%|████████ | 232/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.mlp.gate_proj.weight]
Loading weights: 80%|████████ | 233/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.mlp.up_proj.weight]
Loading weights: 80%|████████ | 233/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.mlp.up_proj.weight]
Loading weights: 81%|████████ | 234/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.post_attention_layernorm.weight]
Loading weights: 81%|████████ | 234/290 [00:01<00:00, 266.25it/s, Materializing param=model.layers.19.post_attention_layernorm.weight]
Loading weights: 81%|████████ | 235/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.post_attention_layernorm.weight]
Loading weights: 81%|████████ | 235/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.k_proj.bias]
Loading weights: 81%|████████ | 235/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.k_proj.bias]
Loading weights: 81%|████████▏ | 236/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.k_proj.weight]
Loading weights: 81%|████████▏ | 236/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.k_proj.weight]
Loading weights: 82%|████████▏ | 237/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.o_proj.weight]
Loading weights: 82%|████████▏ | 237/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.o_proj.weight]
Loading weights: 82%|████████▏ | 238/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.q_proj.bias]
Loading weights: 82%|████████▏ | 238/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.q_proj.bias]
Loading weights: 82%|████████▏ | 239/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.q_proj.weight]
Loading weights: 82%|████████▏ | 239/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.q_proj.weight]
Loading weights: 83%|████████▎ | 240/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.v_proj.bias]
Loading weights: 83%|████████▎ | 240/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.v_proj.bias]
Loading weights: 83%|████████▎ | 241/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.v_proj.weight]
Loading weights: 83%|████████▎ | 241/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.19.self_attn.v_proj.weight]
Loading weights: 83%|████████▎ | 242/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.input_layernorm.weight]
Loading weights: 83%|████████▎ | 242/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.input_layernorm.weight]
Loading weights: 84%|████████▍ | 243/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.mlp.down_proj.weight]
Loading weights: 84%|████████▍ | 243/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.mlp.down_proj.weight]
Loading weights: 84%|████████▍ | 244/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.mlp.gate_proj.weight]
Loading weights: 84%|████████▍ | 244/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.mlp.gate_proj.weight]
Loading weights: 84%|████████▍ | 245/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.mlp.up_proj.weight]
Loading weights: 84%|████████▍ | 245/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.mlp.up_proj.weight]
Loading weights: 85%|████████▍ | 246/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.post_attention_layernorm.weight]
Loading weights: 85%|████████▍ | 246/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.post_attention_layernorm.weight]
Loading weights: 85%|████████▌ | 247/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.k_proj.bias]
Loading weights: 85%|████████▌ | 247/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.k_proj.bias]
Loading weights: 86%|████████▌ | 248/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.k_proj.weight]
Loading weights: 86%|████████▌ | 248/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.k_proj.weight]
Loading weights: 86%|████████▌ | 249/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.o_proj.weight]
Loading weights: 86%|████████▌ | 249/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.o_proj.weight]
Loading weights: 86%|████████▌ | 250/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.q_proj.bias]
Loading weights: 86%|████████▌ | 250/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.q_proj.bias]
Loading weights: 87%|████████▋ | 251/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.q_proj.weight]
Loading weights: 87%|████████▋ | 251/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.q_proj.weight]
Loading weights: 87%|████████▋ | 252/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.v_proj.bias]
Loading weights: 87%|████████▋ | 252/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.v_proj.bias]
Loading weights: 87%|████████▋ | 253/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.v_proj.weight]
Loading weights: 87%|████████▋ | 253/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.20.self_attn.v_proj.weight]
Loading weights: 88%|████████▊ | 254/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.input_layernorm.weight]
Loading weights: 88%|████████▊ | 254/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.input_layernorm.weight]
Loading weights: 88%|████████▊ | 255/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.mlp.down_proj.weight]
Loading weights: 88%|████████▊ | 255/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.mlp.down_proj.weight]
Loading weights: 88%|████████▊ | 256/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.mlp.gate_proj.weight]
Loading weights: 88%|████████▊ | 256/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.mlp.gate_proj.weight]
Loading weights: 89%|████████▊ | 257/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 89%|████████▊ | 257/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 89%|████████▉ | 258/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 89%|████████▉ | 258/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 89%|████████▉ | 259/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.k_proj.bias]
Loading weights: 89%|████████▉ | 259/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.k_proj.bias]
Loading weights: 90%|████████▉ | 260/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.k_proj.weight]
Loading weights: 90%|████████▉ | 260/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.k_proj.weight]
Loading weights: 90%|█████████ | 261/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.o_proj.weight]
Loading weights: 90%|█████████ | 261/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.o_proj.weight]
Loading weights: 90%|█████████ | 262/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.q_proj.bias]
Loading weights: 90%|█████████ | 262/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.q_proj.bias]
Loading weights: 91%|█████████ | 263/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.q_proj.weight]
Loading weights: 91%|█████████ | 263/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.q_proj.weight]
Loading weights: 91%|█████████ | 264/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.v_proj.bias]
Loading weights: 91%|█████████ | 264/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.v_proj.bias]
Loading weights: 91%|█████████▏| 265/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.v_proj.weight]
Loading weights: 91%|█████████▏| 265/290 [00:01<00:00, 287.39it/s, Materializing param=model.layers.21.self_attn.v_proj.weight]
Loading weights: 92%|█████████▏| 266/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.21.self_attn.v_proj.weight]
Loading weights: 92%|█████████▏| 266/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.input_layernorm.weight]
Loading weights: 92%|█████████▏| 266/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.input_layernorm.weight]
Loading weights: 92%|█████████▏| 267/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.mlp.down_proj.weight]
Loading weights: 92%|█████████▏| 267/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.mlp.down_proj.weight]
Loading weights: 92%|█████████▏| 268/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.mlp.gate_proj.weight]
Loading weights: 92%|█████████▏| 268/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.mlp.gate_proj.weight]
Loading weights: 93%|█████████▎| 269/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.mlp.up_proj.weight]
Loading weights: 93%|█████████▎| 269/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.mlp.up_proj.weight]
Loading weights: 93%|█████████▎| 270/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 93%|█████████▎| 270/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 93%|█████████▎| 271/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.k_proj.bias]
Loading weights: 93%|█████████▎| 271/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.k_proj.bias]
Loading weights: 94%|█████████▍| 272/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.k_proj.weight]
Loading weights: 94%|█████████▍| 272/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.k_proj.weight]
Loading weights: 94%|█████████▍| 273/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.o_proj.weight]
Loading weights: 94%|█████████▍| 273/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.o_proj.weight]
Loading weights: 94%|█████████▍| 274/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.q_proj.bias]
Loading weights: 94%|█████████▍| 274/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.q_proj.bias]
Loading weights: 95%|█████████▍| 275/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.q_proj.weight]
Loading weights: 95%|█████████▍| 275/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.q_proj.weight]
Loading weights: 95%|█████████▌| 276/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.v_proj.bias]
Loading weights: 95%|█████████▌| 276/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.v_proj.bias]
Loading weights: 96%|█████████▌| 277/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.v_proj.weight]
Loading weights: 96%|█████████▌| 277/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.22.self_attn.v_proj.weight]
Loading weights: 96%|█████████▌| 278/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.input_layernorm.weight]
Loading weights: 96%|█████████▌| 278/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.input_layernorm.weight]
Loading weights: 96%|█████████▌| 279/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.mlp.down_proj.weight]
Loading weights: 96%|█████████▌| 279/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.mlp.down_proj.weight]
Loading weights: 97%|█████████▋| 280/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 97%|█████████▋| 280/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 97%|█████████▋| 281/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 97%|█████████▋| 281/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 97%|█████████▋| 282/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 97%|█████████▋| 282/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 98%|█████████▊| 283/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.k_proj.bias]
Loading weights: 98%|█████████▊| 283/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.k_proj.bias]
Loading weights: 98%|█████████▊| 284/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.k_proj.weight]
Loading weights: 98%|█████████▊| 284/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.k_proj.weight]
Loading weights: 98%|█████████▊| 285/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.o_proj.weight]
Loading weights: 98%|█████████▊| 285/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.o_proj.weight]
Loading weights: 99%|█████████▊| 286/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.q_proj.bias]
Loading weights: 99%|█████████▊| 286/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.q_proj.bias]
Loading weights: 99%|█████████▉| 287/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.q_proj.weight]
Loading weights: 99%|█████████▉| 287/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.q_proj.weight]
Loading weights: 99%|█████████▉| 288/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.v_proj.bias]
Loading weights: 99%|█████████▉| 288/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.v_proj.bias]
Loading weights: 100%|█████████▉| 289/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.v_proj.weight]
Loading weights: 100%|█████████▉| 289/290 [00:01<00:00, 273.64it/s, Materializing param=model.layers.23.self_attn.v_proj.weight]
Loading weights: 100%|██████████| 290/290 [00:01<00:00, 273.64it/s, Materializing param=model.norm.weight]
Loading weights: 100%|██████████| 290/290 [00:01<00:00, 273.64it/s, Materializing param=model.norm.weight]
Loading weights: 100%|██████████| 290/290 [00:01<00:00, 197.37it/s, Materializing param=model.norm.weight]
unsloth/qwen2.5-0.5b-instruct-unsloth-bnb-4bit does not have a padding token! Will use pad_token = <|PAD_TOKEN|>.
Unsloth 2026.3.3 patched 24 layers with 24 QKV layers, 24 O layers and 24 MLP layers.
Warning: `huggingface-cli` is deprecated and no longer works. Use `hf` instead.

Hint: `hf` is already installed! Use it directly.

Hint: Examples:
hf auth login
hf download unsloth/gemma-4-31B-it-GGUF
hf upload my-cool-model . .
hf models ls --search "gemma"
hf repos ls --format json
hf jobs run python:3.12 python -c 'print("Hello!")'
hf --help

INFO 04-13 02:20:55 [model.py:531] Resolved architecture: Qwen2ForCausalLM
INFO 04-13 02:20:55 [model.py:1554] Using max model len 32768
INFO 04-13 02:20:55 [scheduler.py:231] Chunked prefill is enabled with max_num_batched_tokens=2048.
INFO 04-13 02:20:55 [vllm.py:747] Asynchronous scheduling is enabled.
WARNING 04-13 02:20:57 [system_utils.py:152] We must use the `spawn` multiprocessing start method. Overriding VLLM_WORKER_MULTIPROC_METHOD to 'spawn'. See https://docs.vllm.ai/en/latest/usage/troubleshooting.html#python-multiprocessing for more information. Reasons: CUDA is initialized
Skipping import of cpp extensions due to incompatible torch version 2.10.0+cu128 for torchao version 0.15.0 Please see https://github.com/pytorch/ao/issues/2919 for more info
/usr/local/lib/python3.12/dist-packages/art/__init__.py:37: UserWarning: WARNING: Unsloth should be imported before [transformers] to ensure all optimizations are applied. Your code may run slower or encounter memory issues without these optimizations.
Please restructure your imports with 'import unsloth' at the top of your file.
import unsloth # noqa: F401
🦥 Unsloth: Will patch your computer to enable 2x faster free finetuning.
🦥 Unsloth Zoo will now patch everything to make training faster!
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:21 [core.py:101] Initializing a V1 LLM engine (v0.17.0+art1) with config: model='Qwen/Qwen2.5-0.5B-Instruct', speculative_config=None, tokenizer='Qwen/Qwen2.5-0.5B-Instruct', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, tokenizer_revision=None, trust_remote_code=False, dtype=torch.bfloat16, max_seq_len=32768, download_dir=None, load_format=auto, tensor_parallel_size=1, pipeline_parallel_size=1, data_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, enable_return_routed_experts=False, kv_cache_dtype=auto, device_config=cuda, structured_outputs_config=StructuredOutputsConfig(backend='auto', disable_fallback=False, disable_any_whitespace=False, disable_additional_properties=False, reasoning_parser='', reasoning_parser_plugin='', enable_in_reasoning=False), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None, kv_cache_metrics=False, kv_cache_metrics_sample=0.01, cudagraph_metrics=False, enable_layerwise_nvtx_tracing=False, enable_mfu_metrics=False, enable_mm_processor_stats=False, enable_logging_iteration_details=False), seed=0, served_model_name=Qwen/Qwen2.5-0.5B-Instruct, enable_prefix_caching=True, enable_chunked_prefill=True, pooler_config=None, compilation_config={'level': None, 'mode': <CompilationMode.VLLM_COMPILE: 3>, 'debug_dump_path': None, 'cache_dir': '', 'compile_cache_save_format': 'binary', 'backend': 'inductor', 'custom_ops': ['none'], 'splitting_ops': ['vllm::unified_attention', 'vllm::unified_attention_with_output', 'vllm::unified_mla_attention', 'vllm::unified_mla_attention_with_output', 'vllm::mamba_mixer2', 'vllm::mamba_mixer', 'vllm::short_conv', 'vllm::linear_attention', 'vllm::plamo2_mamba_mixer', 'vllm::gdn_attention_core', 'vllm::kda_attention', 'vllm::sparse_attn_indexer', 'vllm::rocm_aiter_sparse_attn_indexer', 'vllm::unified_kv_cache_update', 'vllm::unified_mla_kv_cache_update'], 'compile_mm_encoder': False, 'compile_sizes': [], 'compile_ranges_split_points': [2048], 'inductor_compile_config': {'enable_auto_functionalized_v2': False, 'combo_kernels': True, 'benchmark_combo_kernel': True}, 'inductor_passes': {}, 'cudagraph_mode': <CUDAGraphMode.FULL_AND_PIECEWISE: (2, 1)>, 'cudagraph_num_of_warmups': 1, 'cudagraph_capture_sizes': [1, 2, 4, 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256], 'cudagraph_copy_inputs': False, 'cudagraph_specialize_lora': True, 'use_inductor_graph_partition': False, 'pass_config': {'fuse_norm_quant': False, 'fuse_act_quant': False, 'fuse_attn_quant': False, 'enable_sp': False, 'fuse_gemm_comms': False, 'fuse_allreduce_rms': False}, 'max_cudagraph_capture_size': 256, 'dynamic_shapes_config': {'type': <DynamicShapesType.BACKED: 'backed'>, 'evaluate_guards': False, 'assume_32_bit_indexing': False}, 'local_cache_dir': None, 'fast_moe_cold_start': True, 'static_all_moe_layers': []}
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:21 [worker_base.py:283] Injected <class 'art.vllm.engine.WorkerExtension'> into <class 'vllm.v1.worker.gpu_worker.Worker'> for extended collective_rpc calls ['run', 'time']
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:21 [parallel_state.py:1393] world_size=1 rank=0 local_rank=0 distributed_init_method=tcp://172.21.0.2:53693 backend=nccl
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:21 [parallel_state.py:1715] rank 0 in world size 1 is assigned as DP rank 0, PP rank 0, PCP rank 0, TP rank 0, EP rank N/A, EPLB rank N/A
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:22 [base.py:106] Offloader set to NoopOffloader
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:22 [gpu_model_runner.py:4255] Starting to load model Qwen/Qwen2.5-0.5B-Instruct...
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:23 [cuda.py:405] Using FLASH_ATTN attention backend out of potential backends: ['FLASH_ATTN', 'FLASHINFER', 'TRITON_ATTN', 'FLEX_ATTENTION'].
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:23 [flash_attn.py:587] Using FlashAttention version 2
(EngineCore_DP0 pid=13597) <frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
(EngineCore_DP0 pid=13597) <frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:24 [weight_utils.py:601] No model.safetensors.index.json found in remote.
(EngineCore_DP0 pid=13597)
Loading safetensors checkpoint shards: 0% Completed | 0/1 [00:00<?, ?it/s]
(EngineCore_DP0 pid=13597)
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:01<00:00, 1.94s/it]
(EngineCore_DP0 pid=13597)
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:01<00:00, 1.94s/it]
(EngineCore_DP0 pid=13597)
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:26 [default_loader.py:293] Loading weights took 1.94 seconds
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:26 [punica_selector.py:20] Using PunicaWrapperGPU.
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:27 [gpu_model_runner.py:4338] Model loading took 0.96 GiB memory and 3.730793 seconds
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:42 [decorators.py:465] Directly load AOT compilation from path /root/.cache/vllm/torch_compile_cache/torch_aot_compile/19f16ef5be162d523fe85c0ed27f944cf1ccd27d08e2ae363d4b7c12b35022cc/rank_0_0/model
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:43 [backends.py:916] Using cache directory: /root/.cache/vllm/torch_compile_cache/d97828e2e7/rank_0_0/backbone for vLLM's torch.compile
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:43 [backends.py:976] Dynamo bytecode transform time: 2.93 s
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:45 [backends.py:266] Directly load the compiled graph(s) for compile range (1, 2048) from the cache, took 1.415 s
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:45 [monitor.py:35] torch.compile takes 5.21 s in total
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:46 [gpu_worker.py:424] Available KV cache memory: 70.01 GiB
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:46 [kv_cache_utils.py:1314] GPU KV cache size: 6,117,600 tokens
(EngineCore_DP0 pid=13597) INFO 04-13 02:21:46 [kv_cache_utils.py:1319] Maximum concurrency for 32,768 tokens per request: 186.69x
(EngineCore_DP0 pid=13597)
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 0%| | 0/70 [00:00<?, ?it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 1%|▏ | 1/70 [00:12<14:38, 12.74s/it](EngineCore_DP0 pid=13597) WARNING 04-13 02:22:01 [utils.py:268] Using default LoRA kernel configs
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 3%|▎ | 2/70 [00:12<06:03, 5.35s/it]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 6%|▌ | 4/70 [00:13<02:14, 2.04s/it]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 9%|▊ | 6/70 [00:13<01:10, 1.10s/it]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 11%|█▏ | 8/70 [00:13<00:42, 1.46it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 14%|█▍ | 10/70 [00:13<00:27, 2.20it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 17%|█▋ | 12/70 [00:13<00:18, 3.13it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 20%|██ | 14/70 [00:13<00:13, 4.27it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 23%|██▎ | 16/70 [00:13<00:09, 5.59it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 26%|██▌ | 18/70 [00:13<00:07, 7.06it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 29%|██▊ | 20/70 [00:14<00:05, 8.59it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 31%|███▏ | 22/70 [00:14<00:04, 9.99it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 34%|███▍ | 24/70 [00:14<00:04, 11.41it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 37%|███▋ | 26/70 [00:14<00:03, 12.69it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 40%|████ | 28/70 [00:14<00:03, 13.76it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 43%|████▎ | 30/70 [00:14<00:02, 14.56it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 46%|████▌ | 32/70 [00:14<00:02, 15.26it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 49%|████▊ | 34/70 [00:14<00:02, 15.86it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 51%|█████▏ | 36/70 [00:14<00:02, 15.83it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 54%|█████▍ | 38/70 [00:15<00:02, 15.92it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 57%|█████▋ | 40/70 [00:15<00:01, 16.33it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 60%|██████ | 42/70 [00:15<00:01, 16.45it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 63%|██████▎ | 44/70 [00:15<00:01, 16.65it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 66%|██████▌ | 46/70 [00:15<00:01, 16.50it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 69%|██████▊ | 48/70 [00:15<00:01, 16.71it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 71%|███████▏ | 50/70 [00:15<00:01, 16.90it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 74%|███████▍ | 52/70 [00:15<00:01, 17.13it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 77%|███████▋ | 54/70 [00:16<00:00, 17.37it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 80%|████████ | 56/70 [00:16<00:00, 17.48it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 83%|████████▎ | 58/70 [00:16<00:00, 16.62it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 86%|████████▌ | 60/70 [00:16<00:00, 16.78it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 89%|████████▊ | 62/70 [00:16<00:00, 16.93it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 91%|█████████▏| 64/70 [00:16<00:00, 17.05it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 94%|█████████▍| 66/70 [00:16<00:00, 17.09it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 97%|█████████▋| 68/70 [00:16<00:00, 17.07it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 100%|██████████| 70/70 [00:17<00:00, 15.27it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 100%|██████████| 70/70 [00:17<00:00, 4.11it/s]
(EngineCore_DP0 pid=13597)
Capturing CUDA graphs (decode, FULL): 0%| | 0/38 [00:00<?, ?it/s]
Capturing CUDA graphs (decode, FULL): 3%|▎ | 1/38 [00:12<07:56, 12.88s/it]
Capturing CUDA graphs (decode, FULL): 11%|█ | 4/38 [00:13<01:24, 2.48s/it]
Capturing CUDA graphs (decode, FULL): 16%|█▌ | 6/38 [00:13<00:45, 1.42s/it]
Capturing CUDA graphs (decode, FULL): 21%|██ | 8/38 [00:13<00:26, 1.11it/s]
Capturing CUDA graphs (decode, FULL): 26%|██▋ | 10/38 [00:13<00:16, 1.66it/s]
Capturing CUDA graphs (decode, FULL): 32%|███▏ | 12/38 [00:13<00:10, 2.40it/s]
Capturing CUDA graphs (decode, FULL): 37%|███▋ | 14/38 [00:13<00:07, 3.34it/s]
Capturing CUDA graphs (decode, FULL): 42%|████▏ | 16/38 [00:13<00:04, 4.50it/s]
Capturing CUDA graphs (decode, FULL): 47%|████▋ | 18/38 [00:13<00:03, 5.88it/s]
Capturing CUDA graphs (decode, FULL): 53%|█████▎ | 20/38 [00:13<00:02, 7.44it/s]
Capturing CUDA graphs (decode, FULL): 58%|█████▊ | 22/38 [00:14<00:01, 9.05it/s]
Capturing CUDA graphs (decode, FULL): 63%|██████▎ | 24/38 [00:14<00:01, 10.63it/s]
Capturing CUDA graphs (decode, FULL): 68%|██████▊ | 26/38 [00:14<00:00, 12.06it/s]
Capturing CUDA graphs (decode, FULL): 74%|███████▎ | 28/38 [00:14<00:00, 13.30it/s]
Capturing CUDA graphs (decode, FULL): 79%|███████▉ | 30/38 [00:14<00:00, 14.35it/s]
Capturing CUDA graphs (decode, FULL): 84%|████████▍ | 32/38 [00:14<00:00, 15.15it/s]
Capturing CUDA graphs (decode, FULL): 89%|████████▉ | 34/38 [00:14<00:00, 15.84it/s]
Capturing CUDA graphs (decode, FULL): 95%|█████████▍| 36/38 [00:14<00:00, 16.25it/s]
Capturing CUDA graphs (decode, FULL): 100%|██████████| 38/38 [00:14<00:00, 16.42it/s]
Capturing CUDA graphs (decode, FULL): 100%|██████████| 38/38 [00:14<00:00, 2.54it/s]
(EngineCore_DP0 pid=13597) INFO 04-13 02:22:21 [gpu_model_runner.py:5360] Graph capturing finished in 33 secs, took 0.65 GiB
(EngineCore_DP0 pid=13597) INFO 04-13 02:22:34 [core.py:282] init engine (profile, create kv cache, warmup model) took 66.79 seconds
(EngineCore_DP0 pid=13597) INFO 04-13 02:22:37 [vllm.py:747] Asynchronous scheduling is enabled.
Starting from step 0
============================================================
Step 1/50
============================================================
step 1: 0%| | 0/32 [00:00<?, ?it/s]
step 1: 3%|▎ | 1/32 [00:01<00:48, 1.57s/it]
step 1: 3%|▎ | 1/32 [00:01<00:48, 1.57s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=44]
step 1: 6%|▋ | 2/32 [00:01<00:22, 1.33it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=44]
step 1: 6%|▋ | 2/32 [00:01<00:22, 1.33it/s, reward=0.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=33]
step 1: 9%|▉ | 3/32 [00:01<00:21, 1.33it/s, reward=1.67, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=29.7]
step 1: 12%|█▎ | 4/32 [00:01<00:21, 1.33it/s, reward=2.25, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=28]
step 1: 16%|█▌ | 5/32 [00:01<00:20, 1.33it/s, reward=1.4, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=26.9]
step 1: 19%|█▉ | 6/32 [00:01<00:19, 1.33it/s, reward=1.5, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=26.2]
step 1: 22%|██▏ | 7/32 [00:01<00:18, 1.33it/s, reward=1.57, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=25.6]
step 1: 25%|██▌ | 8/32 [00:01<00:18, 1.33it/s, reward=1.62, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=25.2]
step 1: 28%|██▊ | 9/32 [00:01<00:17, 1.33it/s, reward=1.67, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=25]
step 1: 31%|███▏ | 10/32 [00:01<00:16, 1.33it/s, reward=1.3, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=25.4]
step 1: 34%|███▍ | 11/32 [00:01<00:02, 9.95it/s, reward=1.3, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=25.4]
step 1: 34%|███▍ | 11/32 [00:01<00:02, 9.95it/s, reward=1.36, num_turns=1.91, num_tools=0.909, failed=0, completion_tokens=26.2]
step 1: 38%|███▊ | 12/32 [00:01<00:02, 9.95it/s, reward=1.33, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=26.7]
step 1: 41%|████ | 13/32 [00:01<00:01, 9.95it/s, reward=1.08, num_turns=1.92, num_tools=0.923, failed=0, completion_tokens=27.2]
step 1: 44%|████▍ | 14/32 [00:01<00:01, 9.95it/s, reward=1.12, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=27.4]
step 1: 47%|████▋ | 15/32 [00:01<00:01, 13.46it/s, reward=1.12, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=27.4]
step 1: 47%|████▋ | 15/32 [00:01<00:01, 13.46it/s, reward=0.911, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=28.2]
step 1: 50%|█████ | 16/32 [00:01<00:01, 13.46it/s, reward=0.729, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=28.9]
step 1: 53%|█████▎ | 17/32 [00:02<00:01, 13.46it/s, reward=0.922, num_turns=1.94, num_tools=0.941, failed=0, completion_tokens=28.5]
step 1: 56%|█████▋ | 18/32 [00:02<00:01, 13.46it/s, reward=0.759, num_turns=1.94, num_tools=0.944, failed=0, completion_tokens=29.8]
step 1: 59%|█████▉ | 19/32 [00:02<00:00, 16.97it/s, reward=0.759, num_turns=1.94, num_tools=0.944, failed=0, completion_tokens=29.8]
step 1: 59%|█████▉ | 19/32 [00:02<00:00, 16.97it/s, reward=0.772, num_turns=1.95, num_tools=0.947, failed=0, completion_tokens=29.6]
step 1: 62%|██████▎ | 20/32 [00:02<00:00, 16.97it/s, reward=0.783, num_turns=1.95, num_tools=0.95, failed=0, completion_tokens=30.7]
step 1: 66%|██████▌ | 21/32 [00:02<00:00, 16.97it/s, reward=0.651, num_turns=1.95, num_tools=0.952, failed=0, completion_tokens=31.9]
step 1: 69%|██████▉ | 22/32 [00:02<00:00, 16.97it/s, reward=0.53, num_turns=1.95, num_tools=0.955, failed=0, completion_tokens=32.5]
step 1: 72%|███████▏ | 23/32 [00:02<00:00, 17.29it/s, reward=0.53, num_turns=1.95, num_tools=0.955, failed=0, completion_tokens=32.5]
step 1: 72%|███████▏ | 23/32 [00:02<00:00, 17.29it/s, reward=0.42, num_turns=1.96, num_tools=0.957, failed=0, completion_tokens=34.3]
step 1: 75%|███████▌ | 24/32 [00:02<00:00, 17.29it/s, reward=0.319, num_turns=1.96, num_tools=0.958, failed=0, completion_tokens=36.3]
step 1: 78%|███████▊ | 25/32 [00:02<00:00, 17.29it/s, reward=0.373, num_turns=1.96, num_tools=0.96, failed=0, completion_tokens=36.2]
step 1: 81%|████████▏ | 26/32 [00:02<00:00, 15.03it/s, reward=0.373, num_turns=1.96, num_tools=0.96, failed=0, completion_tokens=36.2]
step 1: 81%|████████▏ | 26/32 [00:02<00:00, 15.03it/s, reward=0.282, num_turns=1.96, num_tools=0.962, failed=0, completion_tokens=38.5]
step 1: 84%|████████▍ | 27/32 [00:03<00:00, 15.03it/s, reward=0.16, num_turns=1.93, num_tools=0.926, failed=0, completion_tokens=50.6]
step 1: 88%|████████▊ | 28/32 [00:03<00:00, 15.03it/s, reward=0.0833, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=56.6]
step 1: 91%|█████████ | 29/32 [00:03<00:00, 6.02it/s, reward=0.0833, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=56.6]
step 1: 91%|█████████ | 29/32 [00:03<00:00, 6.02it/s, reward=-0.023, num_turns=1.9, num_tools=0.897, failed=0, completion_tokens=72.3]
step 1: 94%|█████████▍| 30/32 [00:04<00:00, 6.02it/s, reward=-0.0889, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=78.7]
step 1: 97%|█████████▋| 31/32 [00:04<00:00, 6.05it/s, reward=-0.0889, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=78.7]
step 1: 97%|█████████▋| 31/32 [00:04<00:00, 6.05it/s, reward=-0.151, num_turns=1.9, num_tools=0.903, failed=0, completion_tokens=84.9]
step 1: 100%|██████████| 32/32 [00:04<00:00, 6.05it/s, reward=-0.208, num_turns=1.91, num_tools=0.906, failed=0, completion_tokens=90.8]
step 1: 100%|██████████| 32/32 [00:04<00:00, 7.54it/s, reward=-0.208, num_turns=1.91, num_tools=0.906, failed=0, completion_tokens=90.8]
group 0: mean=-1.25 std=1.639 min=-3.0 max=+2.0 | What is the distance from Earth to the Sun in km i
group 1: mean=-0.71 std=1.679 min=-2.0 max=+1.7 | What is the population of India divided by its are
group 2: mean=-1.88 std=1.166 min=-3.0 max=+1.0 | How old was Einstein in 2020?
group 3: mean=+3.00 std=1.000 min=+2.0 max=+4.0 | What is 567 times 18?
Avg reward: -0.208 | Avg tools/rollout: 0.9 | groups with variance: 4/4
"./.art/rl-tool-use/models/qwen-0.5b-tool-agent/history.jsonl" not found
Packed 32 trajectories into 4 sequences of length 2048
train: 0%| | 0/4 [00:00<?, ?it/s]The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None}.
==((====))== Unsloth - 2x faster free finetuning | Num GPUs used = 1
\\ /| Num examples = 10,000,000 | Num Epochs = 3 | Total steps = 30,000,000
O^O/ \_/ \ Batch size per device = 2 | Gradient accumulation steps = 1
\ / Data Parallel GPUs = 1 | Total batch size (2 x 1 x 1) = 2
"-____-" Trainable parameters = 4,399,104 of 498,431,872 (0.88% trained)
train: 25%|██▌ | 1/4 [00:10<00:30, 10.23s/it]
train: 25%|██▌ | 1/4 [00:10<00:30, 10.23s/it, loss/train=-1.08, loss/grad_norm=0.826, loss/learning_rate=5e-5, loss/entropy=1.5]
train: 50%|█████ | 2/4 [00:10<00:08, 4.43s/it, loss/train=-1.08, loss/grad_norm=0.826, loss/learning_rate=5e-5, loss/entropy=1.5]
train: 50%|█████ | 2/4 [00:10<00:08, 4.43s/it, loss/train=-0.388, loss/grad_norm=2.59, loss/learning_rate=5e-5, loss/entropy=2.01]
train: 75%|███████▌ | 3/4 [00:10<00:02, 2.57s/it, loss/train=-0.388, loss/grad_norm=2.59, loss/learning_rate=5e-5, loss/entropy=2.01]
train: 75%|███████▌ | 3/4 [00:10<00:02, 2.57s/it, loss/train=-1.22, loss/grad_norm=0.875, loss/learning_rate=5e-5, loss/entropy=1.48]
train: 100%|██████████| 4/4 [00:11<00:00, 1.70s/it, loss/train=-1.22, loss/grad_norm=0.875, loss/learning_rate=5e-5, loss/entropy=1.48]
train: 100%|██████████| 4/4 [00:11<00:00, 1.70s/it, loss/train=-0.368, loss/grad_norm=1.93, loss/learning_rate=5e-5, loss/entropy=1.27](APIServer pid=12946) Adapters before cleanup: ['default']
(APIServer pid=12946) Keeping active adapter(s): ['default']
(APIServer pid=12946) Adapters after cleanup: ['default']
train: 100%|██████████| 4/4 [00:39<00:00, 9.89s/it, loss/train=-0.368, loss/grad_norm=1.93, loss/learning_rate=5e-5, loss/entropy=1.27]
============================================================
Step 2/50
============================================================
step 2: 0%| | 0/32 [00:00<?, ?it/s]
step 2: 3%|▎ | 1/32 [00:01<00:34, 1.12s/it]
step 2: 3%|▎ | 1/32 [00:01<00:34, 1.12s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=36]
step 2: 6%|▋ | 2/32 [00:01<00:33, 1.12s/it, reward=-2.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=27.8]
step 2: 9%|▉ | 3/32 [00:01<00:32, 1.12s/it, reward=-1, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=25.5]
step 2: 12%|█▎ | 4/32 [00:01<00:31, 1.12s/it, reward=-0.25, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=24.4]
step 2: 16%|█▌ | 5/32 [00:01<00:30, 1.12s/it, reward=0.2, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=23.7]
step 2: 19%|█▉ | 6/32 [00:01<00:04, 6.33it/s, reward=0.2, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=23.7]
step 2: 19%|█▉ | 6/32 [00:01<00:04, 6.33it/s, reward=0.5, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=23.2]
step 2: 22%|██▏ | 7/32 [00:01<00:03, 6.33it/s, reward=0.714, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=22.9]
step 2: 25%|██▌ | 8/32 [00:01<00:03, 6.33it/s, reward=0.875, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=22.7]
step 2: 28%|██▊ | 9/32 [00:01<00:03, 6.33it/s, reward=1, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=22.5]
step 2: 31%|███▏ | 10/32 [00:01<00:03, 6.33it/s, reward=1.3, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=23]
step 2: 34%|███▍ | 11/32 [00:01<00:03, 6.33it/s, reward=1.55, num_turns=1.91, num_tools=0.909, failed=0, completion_tokens=23.4]
step 2: 38%|███▊ | 12/32 [00:01<00:03, 6.33it/s, reward=1.75, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=23.8]
step 2: 41%|████ | 13/32 [00:01<00:03, 6.33it/s, reward=1.92, num_turns=1.92, num_tools=0.923, failed=0, completion_tokens=24.2]
step 2: 44%|████▍ | 14/32 [00:01<00:02, 6.33it/s, reward=2.07, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=24.5]
step 2: 47%|████▋ | 15/32 [00:01<00:02, 6.33it/s, reward=2.2, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=24.7]
step 2: 50%|█████ | 16/32 [00:01<00:02, 6.33it/s, reward=2.31, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=24.9]
step 2: 53%|█████▎ | 17/32 [00:01<00:02, 6.33it/s, reward=2.06, num_turns=1.94, num_tools=0.941, failed=0, completion_tokens=24.6]
step 2: 56%|█████▋ | 18/32 [00:01<00:02, 6.33it/s, reward=1.83, num_turns=1.94, num_tools=0.944, failed=0, completion_tokens=25.3]
step 2: 59%|█████▉ | 19/32 [00:01<00:02, 6.33it/s, reward=1.63, num_turns=1.95, num_tools=0.947, failed=0, completion_tokens=25.7]
step 2: 62%|██████▎ | 20/32 [00:01<00:00, 24.63it/s, reward=1.63, num_turns=1.95, num_tools=0.947, failed=0, completion_tokens=25.7]
step 2: 62%|██████▎ | 20/32 [00:01<00:00, 24.63it/s, reward=1.45, num_turns=1.95, num_tools=0.95, failed=0, completion_tokens=26]
step 2: 66%|██████▌ | 21/32 [00:01<00:00, 24.63it/s, reward=1.29, num_turns=1.95, num_tools=0.952, failed=0, completion_tokens=26.3]
step 2: 69%|██████▉ | 22/32 [00:01<00:00, 24.63it/s, reward=1.41, num_turns=1.95, num_tools=0.955, failed=0, completion_tokens=26.2]
step 2: 72%|███████▏ | 23/32 [00:01<00:00, 24.63it/s, reward=1.52, num_turns=1.96, num_tools=0.957, failed=0, completion_tokens=26.2]
step 2: 75%|███████▌ | 24/32 [00:01<00:00, 24.63it/s, reward=1.38, num_turns=1.96, num_tools=0.958, failed=0, completion_tokens=26.1]
step 2: 78%|███████▊ | 25/32 [00:01<00:00, 24.63it/s, reward=1.4, num_turns=1.96, num_tools=0.96, failed=0, completion_tokens=25.9]
step 2: 81%|████████▏ | 26/32 [00:01<00:00, 24.63it/s, reward=1.5, num_turns=1.96, num_tools=0.962, failed=0, completion_tokens=26]
step 2: 84%|████████▍ | 27/32 [00:01<00:00, 24.63it/s, reward=1.59, num_turns=1.96, num_tools=0.963, failed=0, completion_tokens=26.1]
step 2: 88%|████████▊ | 28/32 [00:01<00:00, 33.31it/s, reward=1.59, num_turns=1.96, num_tools=0.963, failed=0, completion_tokens=26.1]
step 2: 88%|████████▊ | 28/32 [00:01<00:00, 33.31it/s, reward=1.68, num_turns=1.96, num_tools=0.964, failed=0, completion_tokens=26.1]
step 2: 91%|█████████ | 29/32 [00:01<00:00, 33.31it/s, reward=1.76, num_turns=1.97, num_tools=0.966, failed=0, completion_tokens=26.2]
step 2: 94%|█████████▍| 30/32 [00:01<00:00, 33.31it/s, reward=1.83, num_turns=1.97, num_tools=0.967, failed=0, completion_tokens=26.3]
step 2: 97%|█████████▋| 31/32 [00:01<00:00, 33.31it/s, reward=1.9, num_turns=1.97, num_tools=0.968, failed=0, completion_tokens=26.3]
step 2: 100%|██████████| 32/32 [00:01<00:00, 33.31it/s, reward=1.97, num_turns=1.97, num_tools=0.969, failed=0, completion_tokens=26.4]
step 2: 100%|██████████| 32/32 [00:01<00:00, 22.07it/s, reward=1.97, num_turns=1.97, num_tools=0.969, failed=0, completion_tokens=26.4]
group 0: mean=+2.00 std=0.000 min=+2.0 max=+2.0 | What is 115 minus 94?
group 1: mean=+4.00 std=0.000 min=+4.0 max=+4.0 | Convert 33 kg to lbs.
group 2: mean=+4.00 std=0.000 min=+4.0 max=+4.0 | Convert 11 kg to lbs.
group 3: mean=-2.12 std=0.331 min=-3.0 max=-2.0 | What is the boiling point of water?
Avg reward: 1.969 | Avg tools/rollout: 1.0 | groups with variance: 1/4
No "val/reward" metric found in history
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0000
Packed 8 trajectories into 1 sequences of length 2048
train: 0%| | 0/1 [00:00<?, ?it/s]
train: 100%|██████████| 1/1 [00:01<00:00, 1.86s/it]
train: 100%|██████████| 1/1 [00:01<00:00, 1.86s/it, loss/train=-0.0999, loss/grad_norm=3, loss/learning_rate=5e-5, loss/entropy=0.884](APIServer pid=12946) Adapters before cleanup: ['default']
(APIServer pid=12946) Keeping active adapter(s): ['default']
(APIServer pid=12946) Adapters after cleanup: ['default']
train: 100%|██████████| 1/1 [00:30<00:00, 30.06s/it, loss/train=-0.0999, loss/grad_norm=3, loss/learning_rate=5e-5, loss/entropy=0.884]
============================================================
Step 3/50
============================================================
step 3: 0%| | 0/32 [00:00<?, ?it/s]
step 3: 3%|▎ | 1/32 [00:01<00:34, 1.13s/it]
step 3: 3%|▎ | 1/32 [00:01<00:34, 1.13s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=22]
step 3: 6%|▋ | 2/32 [00:01<00:18, 1.64it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=22]
step 3: 6%|▋ | 2/32 [00:01<00:18, 1.64it/s, reward=-0.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=21.8]
step 3: 9%|▉ | 3/32 [00:01<00:17, 1.64it/s, reward=0.333, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=21.7]
step 3: 12%|█▎ | 4/32 [00:01<00:17, 1.64it/s, reward=-0.25, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=22.2]
step 3: 16%|█▌ | 5/32 [00:01<00:16, 1.64it/s, reward=-0.8, num_turns=1.6, num_tools=0.6, failed=0, completion_tokens=31.4]
step 3: 19%|█▉ | 6/32 [00:01<00:15, 1.64it/s, reward=-1, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=30.8]
step 3: 22%|██▏ | 7/32 [00:01<00:15, 1.64it/s, reward=-0.286, num_turns=1.71, num_tools=0.714, failed=0, completion_tokens=30.4]
step 3: 25%|██▌ | 8/32 [00:01<00:14, 1.64it/s, reward=0.25, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=30]
step 3: 28%|██▊ | 9/32 [00:01<00:14, 1.64it/s, reward=0.667, num_turns=1.78, num_tools=0.778, failed=0, completion_tokens=29.7]
step 3: 31%|███▏ | 10/32 [00:01<00:13, 1.64it/s, reward=1, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=29.5]
step 3: 34%|███▍ | 11/32 [00:01<00:12, 1.64it/s, reward=1.21, num_turns=1.82, num_tools=0.818, failed=0, completion_tokens=29.4]
step 3: 38%|███▊ | 12/32 [00:01<00:12, 1.64it/s, reward=1.44, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=29.2]
step 3: 41%|████ | 13/32 [00:01<00:11, 1.64it/s, reward=1.64, num_turns=1.85, num_tools=0.846, failed=0, completion_tokens=29.1]
step 3: 44%|████▍ | 14/32 [00:01<00:01, 15.53it/s, reward=1.64, num_turns=1.85, num_tools=0.846, failed=0, completion_tokens=29.1]
step 3: 44%|████▍ | 14/32 [00:01<00:01, 15.53it/s, reward=1.81, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=29.2]
step 3: 47%|████▋ | 15/32 [00:01<00:01, 15.53it/s, reward=1.82, num_turns=1.87, num_tools=0.867, failed=0, completion_tokens=28.7]
step 3: 50%|█████ | 16/32 [00:01<00:01, 15.53it/s, reward=1.83, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=28.2]
step 3: 53%|█████▎ | 17/32 [00:01<00:00, 15.53it/s, reward=1.84, num_turns=1.88, num_tools=0.882, failed=0, completion_tokens=27.9]
step 3: 56%|█████▋ | 18/32 [00:01<00:00, 15.53it/s, reward=1.85, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=27.5]
step 3: 59%|█████▉ | 19/32 [00:01<00:00, 15.53it/s, reward=1.86, num_turns=1.89, num_tools=0.895, failed=0, completion_tokens=27.2]
step 3: 62%|██████▎ | 20/32 [00:01<00:00, 15.53it/s, reward=1.67, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=27.7]
step 3: 66%|██████▌ | 21/32 [00:01<00:00, 15.53it/s, reward=1.49, num_turns=1.9, num_tools=0.905, failed=0, completion_tokens=28.3]
step 3: 69%|██████▉ | 22/32 [00:01<00:00, 15.53it/s, reward=1.33, num_turns=1.91, num_tools=0.909, failed=0, completion_tokens=28.8]
step 3: 72%|███████▏ | 23/32 [00:01<00:00, 15.53it/s, reward=1.19, num_turns=1.91, num_tools=0.913, failed=0, completion_tokens=29.4]
step 3: 75%|███████▌ | 24/32 [00:01<00:00, 27.08it/s, reward=1.19, num_turns=1.91, num_tools=0.913, failed=0, completion_tokens=29.4]
step 3: 75%|███████▌ | 24/32 [00:01<00:00, 27.08it/s, reward=1.17, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=30.1]
step 3: 78%|███████▊ | 25/32 [00:01<00:00, 27.08it/s, reward=1.04, num_turns=1.92, num_tools=0.92, failed=0, completion_tokens=30]
step 3: 81%|████████▏ | 26/32 [00:01<00:00, 27.08it/s, reward=1.03, num_turns=1.92, num_tools=0.923, failed=0, completion_tokens=30.8]
step 3: 84%|████████▍ | 27/32 [00:01<00:00, 27.08it/s, reward=0.914, num_turns=1.93, num_tools=0.926, failed=0, completion_tokens=31.2]
step 3: 88%|████████▊ | 28/32 [00:01<00:00, 27.08it/s, reward=0.929, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=32.7]
step 3: 91%|█████████ | 29/32 [00:01<00:00, 27.08it/s, reward=0.954, num_turns=1.97, num_tools=0.966, failed=0, completion_tokens=33.1]
step 3: 94%|█████████▍| 30/32 [00:01<00:00, 27.08it/s, reward=0.944, num_turns=1.97, num_tools=0.967, failed=0, completion_tokens=33.6]
step 3: 97%|█████████▋| 31/32 [00:02<00:00, 15.58it/s, reward=0.944, num_turns=1.97, num_tools=0.967, failed=0, completion_tokens=33.6]
step 3: 97%|█████████▋| 31/32 [00:02<00:00, 15.58it/s, reward=0.946, num_turns=1.97, num_tools=0.968, failed=0, completion_tokens=35.9]
step 3: 100%|██████████| 32/32 [00:02<00:00, 15.58it/s, reward=0.854, num_turns=1.97, num_tools=0.969, failed=0, completion_tokens=40.7]
step 3: 100%|██████████| 32/32 [00:02<00:00, 11.47it/s, reward=0.854, num_turns=1.97, num_tools=0.969, failed=0, completion_tokens=40.7]
group 0: mean=+1.50 std=1.323 min=-2.0 max=+2.0 | What is 995 minus 50?
group 1: mean=-0.88 std=1.462 min=-2.0 max=+1.3 | What is Germany's population density in people per
group 2: mean=+3.92 std=0.220 min=+3.3 max=+4.0 | Convert 5 kg to lbs.
group 3: mean=-1.12 std=1.715 min=-3.0 max=+1.7 | What is India's population density in people per s
Avg reward: 0.854 | Avg tools/rollout: 1.0 | groups with variance: 4/4
No "val/reward" metric found in history
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0001
Packed 32 trajectories into 3 sequences of length 2048
train: 0%| | 0/3 [00:00<?, ?it/s]
train: 33%|███▎ | 1/3 [00:02<00:04, 2.03s/it]
train: 33%|███▎ | 1/3 [00:02<00:04, 2.03s/it, loss/train=-0.323, loss/grad_norm=0.562, loss/learning_rate=5e-5, loss/entropy=0.441]
train: 67%|██████▋ | 2/3 [00:02<00:01, 1.03s/it, loss/train=-0.323, loss/grad_norm=0.562, loss/learning_rate=5e-5, loss/entropy=0.441]
train: 67%|██████▋ | 2/3 [00:02<00:01, 1.03s/it, loss/train=-0.0952, loss/grad_norm=3.2, loss/learning_rate=5e-5, loss/entropy=1.27]
train: 100%|██████████| 3/3 [00:02<00:00, 1.42it/s, loss/train=-0.0952, loss/grad_norm=3.2, loss/learning_rate=5e-5, loss/entropy=1.27]
train: 100%|██████████| 3/3 [00:02<00:00, 1.42it/s, loss/train=0.106, loss/grad_norm=1.33, loss/learning_rate=5e-5, loss/entropy=0.686](APIServer pid=12946) Adapters before cleanup: ['default']
(APIServer pid=12946) Keeping active adapter(s): ['default']
(APIServer pid=12946) Adapters after cleanup: ['default']
train: 100%|██████████| 3/3 [00:30<00:00, 10.24s/it, loss/train=0.106, loss/grad_norm=1.33, loss/learning_rate=5e-5, loss/entropy=0.686]
============================================================
Step 4/50
============================================================
step 4: 0%| | 0/32 [00:00<?, ?it/s]
step 4: 3%|▎ | 1/32 [00:01<00:38, 1.25s/it]
step 4: 3%|▎ | 1/32 [00:01<00:38, 1.25s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=21.5]
step 4: 6%|▋ | 2/32 [00:01<00:37, 1.25s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=21.5]
step 4: 9%|▉ | 3/32 [00:01<00:36, 1.25s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=21.7]
step 4: 12%|█▎ | 4/32 [00:01<00:34, 1.25s/it, reward=3.5, num_turns=2, num_tools=1, failed=0, completion_tokens=21.6]
step 4: 16%|█▌ | 5/32 [00:01<00:33, 1.25s/it, reward=3.2, num_turns=2, num_tools=1, failed=0, completion_tokens=21.6]
step 4: 19%|█▉ | 6/32 [00:01<00:32, 1.25s/it, reward=3, num_turns=2, num_tools=1, failed=0, completion_tokens=21.6]
step 4: 22%|██▏ | 7/32 [00:01<00:31, 1.25s/it, reward=2.86, num_turns=2, num_tools=1, failed=0, completion_tokens=21.6]
step 4: 25%|██▌ | 8/32 [00:01<00:29, 1.25s/it, reward=2.75, num_turns=2, num_tools=1, failed=0, completion_tokens=21.6]
step 4: 28%|██▊ | 9/32 [00:01<00:28, 1.25s/it, reward=2.67, num_turns=2, num_tools=1, failed=0, completion_tokens=21.6]
step 4: 31%|███▏ | 10/32 [00:01<00:27, 1.25s/it, reward=2.6, num_turns=2, num_tools=1, failed=0, completion_tokens=21.6]
step 4: 34%|███▍ | 11/32 [00:01<00:26, 1.25s/it, reward=2.55, num_turns=2, num_tools=1, failed=0, completion_tokens=21.5]
step 4: 38%|███▊ | 12/32 [00:01<00:24, 1.25s/it, reward=2.17, num_turns=2, num_tools=1, failed=0, completion_tokens=21.7]
step 4: 41%|████ | 13/32 [00:01<00:23, 1.25s/it, reward=2.12, num_turns=2, num_tools=1, failed=0, completion_tokens=21.9]
step 4: 44%|████▍ | 14/32 [00:01<00:01, 9.43it/s, reward=2.12, num_turns=2, num_tools=1, failed=0, completion_tokens=21.9]
step 4: 44%|████▍ | 14/32 [00:01<00:01, 9.43it/s, reward=1.82, num_turns=2, num_tools=1, failed=0, completion_tokens=22.4]
step 4: 47%|████▋ | 15/32 [00:01<00:01, 9.43it/s, reward=1.9, num_turns=2, num_tools=1, failed=0, completion_tokens=22.6]
step 4: 50%|█████ | 16/32 [00:01<00:01, 9.43it/s, reward=1.97, num_turns=2, num_tools=1, failed=0, completion_tokens=22.8]
step 4: 53%|█████▎ | 17/32 [00:01<00:01, 9.43it/s, reward=2.03, num_turns=2, num_tools=1, failed=0, completion_tokens=23.1]
step 4: 56%|█████▋ | 18/32 [00:01<00:01, 9.43it/s, reward=2.14, num_turns=2, num_tools=1, failed=0, completion_tokens=22.9]
step 4: 59%|█████▉ | 19/32 [00:01<00:01, 9.43it/s, reward=2.24, num_turns=2, num_tools=1, failed=0, completion_tokens=22.8]
step 4: 62%|██████▎ | 20/32 [00:01<00:01, 9.43it/s, reward=2.15, num_turns=2, num_tools=1, failed=0, completion_tokens=23]
step 4: 66%|██████▌ | 21/32 [00:01<00:01, 9.43it/s, reward=2.07, num_turns=2, num_tools=1, failed=0, completion_tokens=23.2]
step 4: 69%|██████▉ | 22/32 [00:01<00:01, 9.43it/s, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=23.5]
step 4: 72%|███████▏ | 23/32 [00:01<00:00, 9.43it/s, reward=1.93, num_turns=2, num_tools=1, failed=0, completion_tokens=23.8]
step 4: 75%|███████▌ | 24/32 [00:01<00:00, 9.43it/s, reward=1.88, num_turns=2, num_tools=1, failed=0, completion_tokens=24.1]
step 4: 78%|███████▊ | 25/32 [00:01<00:00, 9.43it/s, reward=1.82, num_turns=2, num_tools=1, failed=0, completion_tokens=24.5]
step 4: 81%|████████▏ | 26/32 [00:01<00:00, 9.43it/s, reward=1.63, num_turns=1.96, num_tools=0.962, failed=0, completion_tokens=28.7]
step 4: 84%|████████▍ | 27/32 [00:01<00:00, 9.43it/s, reward=1.61, num_turns=1.96, num_tools=0.963, failed=0, completion_tokens=28.7]
step 4: 88%|████████▊ | 28/32 [00:01<00:00, 9.43it/s, reward=1.48, num_turns=1.96, num_tools=0.964, failed=0, completion_tokens=29]
step 4: 91%|█████████ | 29/32 [00:01<00:00, 9.43it/s, reward=1.36, num_turns=1.97, num_tools=0.966, failed=0, completion_tokens=29.3]
step 4: 94%|█████████▍| 30/32 [00:01<00:00, 9.43it/s, reward=1.22, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=32.1]
step 4: 97%|█████████▋| 31/32 [00:01<00:00, 9.43it/s, reward=1.21, num_turns=1.94, num_tools=0.935, failed=0, completion_tokens=33]
step 4: 100%|██████████| 32/32 [00:02<00:00, 22.73it/s, reward=1.21, num_turns=1.94, num_tools=0.935, failed=0, completion_tokens=33]
step 4: 100%|██████████| 32/32 [00:02<00:00, 22.73it/s, reward=1.19, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=35.7]
step 4: 100%|██████████| 32/32 [00:02<00:00, 15.81it/s, reward=1.19, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=35.7]
group 0: mean=-1.04 std=1.531 min=-3.0 max=+1.0 | What is Germany's population density in people per
group 1: mean=+3.62 std=0.484 min=+3.0 max=+4.0 | What's the weather like in London?
group 2: mean=+0.19 std=1.248 min=-3.0 max=+1.5 | What is the temperature in Mumbai in Fahrenheit?
group 3: mean=+2.00 std=0.000 min=+2.0 max=+2.0 | What is 543 plus 96?
Avg reward: 1.193 | Avg tools/rollout: 0.9 | groups with variance: 3/4
No "val/reward" metric found in history
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0002
Packed 24 trajectories into 2 sequences of length 2048
train: 0%| | 0/2 [00:00<?, ?it/s]
train: 50%|█████ | 1/2 [00:01<00:01, 1.97s/it]
train: 50%|█████ | 1/2 [00:01<00:01, 1.97s/it, loss/train=0.0658, loss/grad_norm=1.35, loss/learning_rate=5e-5, loss/entropy=0.857]
train: 100%|██████████| 2/2 [00:02<00:00, 1.02it/s, loss/train=0.0658, loss/grad_norm=1.35, loss/learning_rate=5e-5, loss/entropy=0.857]
train: 100%|██████████| 2/2 [00:02<00:00, 1.02it/s, loss/train=-0.501, loss/grad_norm=1.91, loss/learning_rate=5e-5, loss/entropy=0.515](APIServer pid=12946) Adapters before cleanup: ['default']
(APIServer pid=12946) Keeping active adapter(s): ['default']
(APIServer pid=12946) Adapters after cleanup: ['default']
train: 100%|██████████| 2/2 [00:30<00:00, 15.18s/it, loss/train=-0.501, loss/grad_norm=1.91, loss/learning_rate=5e-5, loss/entropy=0.515]
============================================================
Step 5/50
============================================================
step 5: 0%| | 0/32 [00:00<?, ?it/s]
step 5: 3%|▎ | 1/32 [00:01<00:35, 1.15s/it]
step 5: 3%|▎ | 1/32 [00:01<00:35, 1.15s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=28]
step 5: 6%|▋ | 2/32 [00:01<00:16, 1.82it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=28]
step 5: 6%|▋ | 2/32 [00:01<00:16, 1.82it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=34]
step 5: 9%|▉ | 3/32 [00:01<00:15, 1.82it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=37]
step 5: 12%|█▎ | 4/32 [00:01<00:15, 1.82it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=38.8]
step 5: 16%|█▌ | 5/32 [00:01<00:05, 5.27it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=38.8]
step 5: 16%|█▌ | 5/32 [00:01<00:05, 5.27it/s, reward=-2.8, num_turns=1.2, num_tools=0.2, failed=0, completion_tokens=36.2]
step 5: 19%|█▉ | 6/32 [00:01<00:04, 5.27it/s, reward=-2.83, num_turns=1.17, num_tools=0.167, failed=0, completion_tokens=41.5]
step 5: 22%|██▏ | 7/32 [00:01<00:04, 5.27it/s, reward=-2.71, num_turns=1.29, num_tools=0.286, failed=0, completion_tokens=39.4]
step 5: 25%|██▌ | 8/32 [00:01<00:04, 5.27it/s, reward=-2.62, num_turns=1.38, num_tools=0.375, failed=0, completion_tokens=37.9]
step 5: 28%|██▊ | 9/32 [00:01<00:04, 5.27it/s, reward=-2.56, num_turns=1.44, num_tools=0.444, failed=0, completion_tokens=36.6]
step 5: 31%|███▏ | 10/32 [00:01<00:04, 5.27it/s, reward=-2.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=35.8]
step 5: 34%|███▍ | 11/32 [00:01<00:03, 5.27it/s, reward=-2.45, num_turns=1.55, num_tools=0.545, failed=0, completion_tokens=35.5]
step 5: 38%|███▊ | 12/32 [00:01<00:01, 14.50it/s, reward=-2.45, num_turns=1.55, num_tools=0.545, failed=0, completion_tokens=35.5]
step 5: 38%|███▊ | 12/32 [00:01<00:01, 14.50it/s, reward=-2.42, num_turns=1.58, num_tools=0.583, failed=0, completion_tokens=35.7]
step 5: 41%|████ | 13/32 [00:01<00:01, 14.50it/s, reward=-2.38, num_turns=1.62, num_tools=0.615, failed=0, completion_tokens=35.8]
step 5: 44%|████▍ | 14/32 [00:01<00:01, 14.50it/s, reward=-2.36, num_turns=1.64, num_tools=0.643, failed=0, completion_tokens=36.1]
step 5: 47%|████▋ | 15/32 [00:01<00:01, 14.50it/s, reward=-2.33, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=35.5]
step 5: 50%|█████ | 16/32 [00:01<00:01, 14.50it/s, reward=-2.31, num_turns=1.69, num_tools=0.688, failed=0, completion_tokens=34.7]
step 5: 53%|█████▎ | 17/32 [00:01<00:01, 14.50it/s, reward=-2.29, num_turns=1.71, num_tools=0.706, failed=0, completion_tokens=35.3]
step 5: 56%|█████▋ | 18/32 [00:01<00:00, 21.49it/s, reward=-2.29, num_turns=1.71, num_tools=0.706, failed=0, completion_tokens=35.3]
step 5: 56%|█████▋ | 18/32 [00:01<00:00, 21.49it/s, reward=-2.28, num_turns=1.72, num_tools=0.722, failed=0, completion_tokens=35.9]
step 5: 59%|█████▉ | 19/32 [00:01<00:00, 21.49it/s, reward=-2.05, num_turns=1.74, num_tools=0.737, failed=0, completion_tokens=35.4]
step 5: 62%|██████▎ | 20/32 [00:01<00:00, 21.49it/s, reward=-2.05, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=36.2]
step 5: 66%|██████▌ | 21/32 [00:01<00:00, 21.49it/s, reward=-2.05, num_turns=1.76, num_tools=0.762, failed=0, completion_tokens=37.2]
step 5: 69%|██████▉ | 22/32 [00:01<00:00, 24.26it/s, reward=-2.05, num_turns=1.76, num_tools=0.762, failed=0, completion_tokens=37.2]
step 5: 69%|██████▉ | 22/32 [00:01<00:00, 24.26it/s, reward=-2.05, num_turns=1.77, num_tools=0.773, failed=0, completion_tokens=38.1]
step 5: 72%|███████▏ | 23/32 [00:01<00:00, 24.26it/s, reward=-2.04, num_turns=1.78, num_tools=0.783, failed=0, completion_tokens=37.7]
step 5: 75%|███████▌ | 24/32 [00:01<00:00, 24.26it/s, reward=-2.04, num_turns=1.79, num_tools=0.792, failed=0, completion_tokens=38.2]
step 5: 78%|███████▊ | 25/32 [00:01<00:00, 24.26it/s, reward=-2.04, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=37.8]
step 5: 81%|████████▏ | 26/32 [00:01<00:00, 23.28it/s, reward=-2.04, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=37.8]
step 5: 81%|████████▏ | 26/32 [00:01<00:00, 23.28it/s, reward=-2.04, num_turns=1.81, num_tools=0.808, failed=0, completion_tokens=38.3]
step 5: 84%|████████▍ | 27/32 [00:02<00:00, 23.28it/s, reward=-2.04, num_turns=1.81, num_tools=0.815, failed=0, completion_tokens=40.4]
step 5: 88%|████████▊ | 28/32 [00:02<00:00, 23.28it/s, reward=-2.07, num_turns=1.79, num_tools=0.786, failed=0, completion_tokens=47.9]
step 5: 91%|█████████ | 29/32 [00:03<00:00, 23.28it/s, reward=-2.1, num_turns=1.76, num_tools=0.759, failed=0, completion_tokens=63.7]
step 5: 94%|█████████▍| 30/32 [00:03<00:00, 7.07it/s, reward=-2.1, num_turns=1.76, num_tools=0.759, failed=0, completion_tokens=63.7]
step 5: 94%|█████████▍| 30/32 [00:03<00:00, 7.07it/s, reward=-2.1, num_turns=1.77, num_tools=0.8, failed=0, completion_tokens=70]
step 5: 97%|█████████▋| 31/32 [00:03<00:00, 7.07it/s, reward=-2.1, num_turns=1.77, num_tools=0.806, failed=0, completion_tokens=76.4]
step 5: 100%|██████████| 32/32 [00:03<00:00, 7.07it/s, reward=-2.09, num_turns=1.78, num_tools=0.812, failed=0, completion_tokens=82.7]
step 5: 100%|██████████| 32/32 [00:03<00:00, 8.72it/s, reward=-2.09, num_turns=1.78, num_tools=0.812, failed=0, completion_tokens=82.7]
group 0: mean=-2.00 std=1.581 min=-3.0 max=+2.0 | What is the tallest mountain?
group 1: mean=-2.00 std=0.000 min=-2.0 max=-2.0 | Which is hotter right now, Tokyo or Mumbai?
group 2: mean=-2.25 std=0.433 min=-3.0 max=-2.0 | What is the population of France divided by its ar
group 3: mean=-2.12 std=0.331 min=-3.0 max=-2.0 | What is the GDP of France?
Avg reward: -2.094 | Avg tools/rollout: 0.8 | groups with variance: 3/4
No "val/reward" metric found in history
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0003
Packed 21 trajectories into 2 sequences of length 2048
train: 0%| | 0/2 [00:00<?, ?it/s]
train: 50%|█████ | 1/2 [00:02<00:02, 2.12s/it]
train: 50%|█████ | 1/2 [00:02<00:02, 2.12s/it, loss/train=-1.41, loss/grad_norm=1.82, loss/learning_rate=5e-5, loss/entropy=1.69]
train: 100%|██████████| 2/2 [00:02<00:00, 1.09s/it, loss/train=-1.41, loss/grad_norm=1.82, loss/learning_rate=5e-5, loss/entropy=1.69]
train: 100%|██████████| 2/2 [00:02<00:00, 1.09s/it, loss/train=-0.442, loss/grad_norm=7.41, loss/learning_rate=5e-5, loss/entropy=1.25](APIServer pid=12946) Adapters before cleanup: ['default']
(APIServer pid=12946) Keeping active adapter(s): ['default']
(APIServer pid=12946) Adapters after cleanup: ['default']
train: 100%|██████████| 2/2 [00:30<00:00, 15.43s/it, loss/train=-0.442, loss/grad_norm=7.41, loss/learning_rate=5e-5, loss/entropy=1.25]
Running validation...
validation: 0%| | 0/400 [00:00<?, ?it/s]
validation: 0%| | 1/400 [00:10<1:12:15, 10.87s/it]
validation: 0%| | 1/400 [00:10<1:12:15, 10.87s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=44]
validation: 0%| | 2/400 [00:11<31:36, 4.77s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=44]
validation: 0%| | 2/400 [00:11<31:36, 4.77s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=37]
validation: 1%| | 3/400 [00:11<31:31, 4.77s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=49.3]
validation: 1%| | 4/400 [00:11<31:27, 4.77s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=54]
validation: 1%|▏ | 5/400 [00:11<31:22, 4.77s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=46.2]
validation: 2%|▏ | 6/400 [00:11<07:32, 1.15s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=46.2]
validation: 2%|▏ | 6/400 [00:11<07:32, 1.15s/it, reward=-2.83, num_turns=1.17, num_tools=0.167, failed=0, completion_tokens=42.2]
validation: 2%|▏ | 7/400 [00:11<06:10, 1.06it/s, reward=-2.83, num_turns=1.17, num_tools=0.167, failed=0, completion_tokens=42.2]
validation: 2%|▏ | 7/400 [00:11<06:10, 1.06it/s, reward=-2.86, num_turns=1.14, num_tools=0.143, failed=0, completion_tokens=42.6]
validation: 2%|▏ | 8/400 [00:11<06:09, 1.06it/s, reward=-2.88, num_turns=1.12, num_tools=0.125, failed=0, completion_tokens=48.2]
validation: 2%|▏ | 9/400 [00:12<04:04, 1.60it/s, reward=-2.88, num_turns=1.12, num_tools=0.125, failed=0, completion_tokens=48.2]
validation: 2%|▏ | 9/400 [00:12<04:04, 1.60it/s, reward=-2.89, num_turns=1.11, num_tools=0.111, failed=0, completion_tokens=45.3]
validation: 2%|▎ | 10/400 [00:12<04:03, 1.60it/s, reward=-2.9, num_turns=1.1, num_tools=0.1, failed=0, completion_tokens=43.3]
validation: 3%|▎ | 11/400 [00:12<02:46, 2.34it/s, reward=-2.9, num_turns=1.1, num_tools=0.1, failed=0, completion_tokens=43.3]
validation: 3%|▎ | 11/400 [00:12<02:46, 2.34it/s, reward=-2.91, num_turns=1.09, num_tools=0.0909, failed=0, completion_tokens=42]
validation: 3%|▎ | 12/400 [00:12<02:45, 2.34it/s, reward=-2.92, num_turns=1.08, num_tools=0.0833, failed=0, completion_tokens=40]
validation: 3%|▎ | 13/400 [00:12<02:45, 2.34it/s, reward=-2.38, num_turns=1.15, num_tools=0.154, failed=0, completion_tokens=39.1]
validation: 4%|▎ | 14/400 [00:12<02:45, 2.34it/s, reward=-1.93, num_turns=1.21, num_tools=0.214, failed=0, completion_tokens=38.3]
validation: 4%|▍ | 15/400 [00:12<02:44, 2.34it/s, reward=-2, num_turns=1.2, num_tools=0.2, failed=0, completion_tokens=37.3]
validation: 4%|▍ | 16/400 [00:12<02:44, 2.34it/s, reward=-2.06, num_turns=1.19, num_tools=0.188, failed=0, completion_tokens=38.4]
validation: 4%|▍ | 17/400 [00:12<02:43, 2.34it/s, reward=-2.06, num_turns=1.24, num_tools=0.235, failed=0, completion_tokens=38]
validation: 4%|▍ | 18/400 [00:12<02:43, 2.34it/s, reward=-2.06, num_turns=1.28, num_tools=0.278, failed=0, completion_tokens=37.4]
validation: 5%|▍ | 19/400 [00:12<02:42, 2.34it/s, reward=-2.05, num_turns=1.32, num_tools=0.316, failed=0, completion_tokens=37.1]
validation: 5%|▌ | 20/400 [00:12<02:42, 2.34it/s, reward=-2.1, num_turns=1.3, num_tools=0.3, failed=0, completion_tokens=39.4]
validation: 5%|▌ | 21/400 [00:12<02:42, 2.34it/s, reward=-1.93, num_turns=1.33, num_tools=0.333, failed=0, completion_tokens=39.5]
validation: 6%|▌ | 22/400 [00:12<02:41, 2.34it/s, reward=-1.98, num_turns=1.32, num_tools=0.318, failed=0, completion_tokens=48.6]
validation: 6%|▌ | 23/400 [00:12<02:41, 2.34it/s, reward=-2.02, num_turns=1.3, num_tools=0.304, failed=0, completion_tokens=53.1]
validation: 6%|▌ | 24/400 [00:12<02:40, 2.34it/s, reward=-1.8, num_turns=1.33, num_tools=0.333, failed=0, completion_tokens=52.1]
validation: 6%|▋ | 25/400 [00:12<02:40, 2.34it/s, reward=-1.81, num_turns=1.36, num_tools=0.36, failed=0, completion_tokens=52.2]
validation: 6%|▋ | 26/400 [00:12<02:39, 2.34it/s, reward=-1.81, num_turns=1.38, num_tools=0.385, failed=0, completion_tokens=52.5]
validation: 7%|▋ | 27/400 [00:12<00:40, 9.10it/s, reward=-1.81, num_turns=1.38, num_tools=0.385, failed=0, completion_tokens=52.5]
validation: 7%|▋ | 27/400 [00:12<00:40, 9.10it/s, reward=-1.64, num_turns=1.41, num_tools=0.407, failed=0, completion_tokens=51.5]
validation: 7%|▋ | 28/400 [00:12<00:40, 9.10it/s, reward=-1.65, num_turns=1.43, num_tools=0.429, failed=0, completion_tokens=50.4]
validation: 7%|▋ | 29/400 [00:12<00:40, 9.10it/s, reward=-1.57, num_turns=1.45, num_tools=0.448, failed=0, completion_tokens=49.6]
validation: 8%|▊ | 30/400 [00:12<00:40, 9.10it/s, reward=-1.51, num_turns=1.47, num_tools=0.467, failed=0, completion_tokens=48.9]
validation: 8%|▊ | 31/400 [00:12<00:40, 9.10it/s, reward=-1.52, num_turns=1.48, num_tools=0.484, failed=0, completion_tokens=48.1]
validation: 8%|▊ | 32/400 [00:12<00:40, 9.10it/s, reward=-1.57, num_turns=1.47, num_tools=0.469, failed=0, completion_tokens=47.4][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 8%|▊ | 33/400 [00:12<00:32, 11.43it/s, reward=-1.57, num_turns=1.47, num_tools=0.469, failed=0, completion_tokens=47.4]
validation: 8%|▊ | 33/400 [00:12<00:32, 11.43it/s, reward=-1.61, num_turns=1.45, num_tools=0.455, failed=0.0303, completion_tokens=47.4]
validation: 8%|▊ | 34/400 [00:12<00:32, 11.43it/s, reward=-1.65, num_turns=1.44, num_tools=0.441, failed=0.0588, completion_tokens=47.4]
validation: 9%|▉ | 35/400 [00:12<00:31, 11.43it/s, reward=-1.69, num_turns=1.43, num_tools=0.429, failed=0.0857, completion_tokens=47.4]
validation: 9%|▉ | 36/400 [00:12<00:31, 11.43it/s, reward=-1.73, num_turns=1.42, num_tools=0.417, failed=0.111, completion_tokens=47.4]
validation: 9%|▉ | 37/400 [00:12<00:31, 11.43it/s, reward=-1.76, num_turns=1.41, num_tools=0.405, failed=0.135, completion_tokens=47.4]
validation: 10%|▉ | 38/400 [00:12<00:31, 11.43it/s, reward=-1.79, num_turns=1.39, num_tools=0.395, failed=0.158, completion_tokens=47.4]
validation: 10%|▉ | 39/400 [00:12<00:31, 11.43it/s, reward=-1.82, num_turns=1.38, num_tools=0.385, failed=0.179, completion_tokens=47.4]
validation: 10%|█ | 40/400 [00:12<00:31, 11.43it/s, reward=-1.85, num_turns=1.38, num_tools=0.375, failed=0.2, completion_tokens=47.4]
validation: 10%|█ | 41/400 [00:12<00:31, 11.43it/s, reward=-1.88, num_turns=1.37, num_tools=0.366, failed=0.22, completion_tokens=47.4]
validation: 10%|█ | 42/400 [00:12<00:31, 11.43it/s, reward=-1.91, num_turns=1.36, num_tools=0.357, failed=0.238, completion_tokens=47.4]
validation: 11%|█ | 43/400 [00:12<00:31, 11.43it/s, reward=-1.93, num_turns=1.35, num_tools=0.349, failed=0.256, completion_tokens=47.4]
validation: 11%|█ | 44/400 [00:12<00:31, 11.43it/s, reward=-1.96, num_turns=1.34, num_tools=0.341, failed=0.273, completion_tokens=47.4]
validation: 11%|█▏ | 45/400 [00:12<00:31, 11.43it/s, reward=-1.98, num_turns=1.33, num_tools=0.333, failed=0.289, completion_tokens=47.4]
validation: 12%|█▏ | 46/400 [00:12<00:30, 11.43it/s, reward=-2, num_turns=1.33, num_tools=0.326, failed=0.304, completion_tokens=47.4] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 12%|█▏ | 47/400 [00:13<00:18, 19.01it/s, reward=-2, num_turns=1.33, num_tools=0.326, failed=0.304, completion_tokens=47.4]
validation: 12%|█▏ | 47/400 [00:13<00:18, 19.01it/s, reward=-2.02, num_turns=1.32, num_tools=0.319, failed=0.319, completion_tokens=47.4]
validation: 12%|█▏ | 48/400 [00:13<00:18, 19.01it/s, reward=-2.05, num_turns=1.31, num_tools=0.312, failed=0.333, completion_tokens=47.4]
validation: 12%|█▏ | 49/400 [00:13<00:18, 19.01it/s, reward=-2.06, num_turns=1.31, num_tools=0.306, failed=0.347, completion_tokens=47.4]
validation: 12%|█▎ | 50/400 [00:13<00:18, 19.01it/s, reward=-2.08, num_turns=1.3, num_tools=0.3, failed=0.36, completion_tokens=47.4]
validation: 13%|█▎ | 51/400 [00:13<00:18, 19.01it/s, reward=-2.1, num_turns=1.29, num_tools=0.294, failed=0.373, completion_tokens=47.4]
validation: 13%|█▎ | 52/400 [00:13<00:18, 19.01it/s, reward=-2.12, num_turns=1.29, num_tools=0.288, failed=0.385, completion_tokens=47.4]
validation: 13%|█▎ | 53/400 [00:13<00:18, 19.01it/s, reward=-2.14, num_turns=1.28, num_tools=0.283, failed=0.396, completion_tokens=47.4]
validation: 14%|█▎ | 54/400 [00:13<00:18, 19.01it/s, reward=-2.15, num_turns=1.28, num_tools=0.278, failed=0.407, completion_tokens=47.4]
validation: 14%|█▍ | 55/400 [00:13<00:18, 19.01it/s, reward=-2.17, num_turns=1.27, num_tools=0.273, failed=0.418, completion_tokens=47.4]
validation: 14%|█▍ | 56/400 [00:13<00:18, 19.01it/s, reward=-2.18, num_turns=1.27, num_tools=0.268, failed=0.429, completion_tokens=47.4]
validation: 14%|█▍ | 57/400 [00:13<00:18, 19.01it/s, reward=-2.2, num_turns=1.26, num_tools=0.263, failed=0.439, completion_tokens=47.4]
validation: 14%|█▍ | 58/400 [00:13<00:17, 19.01it/s, reward=-2.21, num_turns=1.26, num_tools=0.259, failed=0.448, completion_tokens=47.4]
validation: 15%|█▍ | 59/400 [00:13<00:17, 19.01it/s, reward=-2.22, num_turns=1.25, num_tools=0.254, failed=0.458, completion_tokens=47.4]
validation: 15%|█▌ | 60/400 [00:13<00:17, 19.01it/s, reward=-2.24, num_turns=1.25, num_tools=0.25, failed=0.467, completion_tokens=47.4]
validation: 15%|█▌ | 61/400 [00:13<00:17, 19.01it/s, reward=-2.25, num_turns=1.25, num_tools=0.246, failed=0.475, completion_tokens=47.4]
validation: 16%|█▌ | 62/400 [00:13<00:17, 19.01it/s, reward=-2.26, num_turns=1.24, num_tools=0.242, failed=0.484, completion_tokens=47.4]
validation: 16%|█▌ | 63/400 [00:13<00:17, 19.01it/s, reward=-2.27, num_turns=1.24, num_tools=0.238, failed=0.492, completion_tokens=47.4]
validation: 16%|█▌ | 64/400 [00:13<00:17, 19.01it/s, reward=-2.28, num_turns=1.23, num_tools=0.234, failed=0.5, completion_tokens=47.4]
validation: 16%|█▋ | 65/400 [00:13<00:17, 19.01it/s, reward=-2.29, num_turns=1.23, num_tools=0.231, failed=0.508, completion_tokens=47.4]
validation: 16%|█▋ | 66/400 [00:13<00:17, 19.01it/s, reward=-2.31, num_turns=1.23, num_tools=0.227, failed=0.515, completion_tokens=47.4]
validation: 17%|█▋ | 67/400 [00:13<00:17, 19.01it/s, reward=-2.32, num_turns=1.22, num_tools=0.224, failed=0.522, completion_tokens=47.4]
validation: 17%|█▋ | 68/400 [00:13<00:17, 19.01it/s, reward=-2.33, num_turns=1.22, num_tools=0.221, failed=0.529, completion_tokens=47.4]
validation: 17%|█▋ | 69/400 [00:13<00:17, 19.01it/s, reward=-2.34, num_turns=1.22, num_tools=0.217, failed=0.536, completion_tokens=47.4]
validation: 18%|█▊ | 70/400 [00:13<00:17, 19.01it/s, reward=-2.35, num_turns=1.21, num_tools=0.214, failed=0.543, completion_tokens=47.4]
validation: 18%|█▊ | 71/400 [00:13<00:17, 19.01it/s, reward=-2.35, num_turns=1.21, num_tools=0.211, failed=0.549, completion_tokens=47.4]
validation: 18%|█▊ | 72/400 [00:13<00:17, 19.01it/s, reward=-2.36, num_turns=1.21, num_tools=0.208, failed=0.556, completion_tokens=47.4]
validation: 18%|█▊ | 73/400 [00:13<00:17, 19.01it/s, reward=-2.37, num_turns=1.21, num_tools=0.205, failed=0.562, completion_tokens=47.4][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 18%|█▊ | 74/400 [00:13<00:17, 19.01it/s, reward=-2.38, num_turns=1.2, num_tools=0.203, failed=0.568, completion_tokens=47.4] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 19%|█▉ | 75/400 [00:13<00:08, 39.70it/s, reward=-2.38, num_turns=1.2, num_tools=0.203, failed=0.568, completion_tokens=47.4]
validation: 19%|█▉ | 75/400 [00:13<00:08, 39.70it/s, reward=-2.39, num_turns=1.2, num_tools=0.2, failed=0.573, completion_tokens=47.4]
validation: 19%|█▉ | 76/400 [00:13<00:08, 39.70it/s, reward=-2.4, num_turns=1.2, num_tools=0.197, failed=0.579, completion_tokens=47.4][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 19%|█▉ | 77/400 [00:13<00:08, 39.70it/s, reward=-2.4, num_turns=1.19, num_tools=0.195, failed=0.584, completion_tokens=47.4]
validation: 20%|█▉ | 78/400 [00:13<00:08, 39.70it/s, reward=-2.41, num_turns=1.19, num_tools=0.192, failed=0.59, completion_tokens=47.4]
validation: 20%|█▉ | 79/400 [00:13<00:08, 39.70it/s, reward=-2.42, num_turns=1.19, num_tools=0.19, failed=0.595, completion_tokens=47.4]
validation: 20%|██ | 80/400 [00:13<00:08, 39.70it/s, reward=-2.43, num_turns=1.19, num_tools=0.188, failed=0.6, completion_tokens=47.4]
validation: 20%|██ | 81/400 [00:13<00:08, 39.70it/s, reward=-2.43, num_turns=1.19, num_tools=0.185, failed=0.605, completion_tokens=47.4]
validation: 20%|██ | 82/400 [00:13<00:08, 39.70it/s, reward=-2.44, num_turns=1.18, num_tools=0.183, failed=0.61, completion_tokens=47.4]
validation: 21%|██ | 83/400 [00:13<00:07, 39.70it/s, reward=-2.45, num_turns=1.18, num_tools=0.181, failed=0.614, completion_tokens=47.4]
validation: 21%|██ | 84/400 [00:13<00:07, 39.70it/s, reward=-2.45, num_turns=1.18, num_tools=0.179, failed=0.619, completion_tokens=47.4]
validation: 21%|██▏ | 85/400 [00:13<00:07, 39.70it/s, reward=-2.46, num_turns=1.18, num_tools=0.176, failed=0.624, completion_tokens=47.4]
validation: 22%|██▏ | 86/400 [00:13<00:07, 39.70it/s, reward=-2.47, num_turns=1.17, num_tools=0.174, failed=0.628, completion_tokens=47.4]
validation: 22%|██▏ | 87/400 [00:13<00:06, 44.77it/s, reward=-2.47, num_turns=1.17, num_tools=0.174, failed=0.628, completion_tokens=47.4]
validation: 22%|██▏ | 87/400 [00:13<00:06, 44.77it/s, reward=-2.46, num_turns=1.18, num_tools=0.184, failed=0.621, completion_tokens=46.6]
validation: 22%|██▏ | 88/400 [00:13<00:06, 44.77it/s, reward=-2.46, num_turns=1.19, num_tools=0.193, failed=0.614, completion_tokens=45.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 22%|██▏ | 89/400 [00:13<00:06, 44.77it/s, reward=-2.46, num_turns=1.19, num_tools=0.191, failed=0.618, completion_tokens=45.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 22%|██▎ | 90/400 [00:13<00:06, 44.77it/s, reward=-2.47, num_turns=1.19, num_tools=0.189, failed=0.622, completion_tokens=45.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 23%|██▎ | 91/400 [00:13<00:06, 44.77it/s, reward=-2.47, num_turns=1.19, num_tools=0.187, failed=0.626, completion_tokens=45.9]
validation: 23%|██▎ | 92/400 [00:13<00:06, 44.77it/s, reward=-2.48, num_turns=1.18, num_tools=0.185, failed=0.63, completion_tokens=45.9] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 23%|██▎ | 93/400 [00:13<00:06, 44.77it/s, reward=-2.49, num_turns=1.18, num_tools=0.183, failed=0.634, completion_tokens=45.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 24%|██▎ | 94/400 [00:13<00:06, 44.77it/s, reward=-2.49, num_turns=1.18, num_tools=0.181, failed=0.638, completion_tokens=45.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 24%|██▍ | 95/400 [00:13<00:06, 44.77it/s, reward=-2.5, num_turns=1.18, num_tools=0.179, failed=0.642, completion_tokens=45.9] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 24%|██▍ | 96/400 [00:13<00:06, 44.77it/s, reward=-2.5, num_turns=1.18, num_tools=0.177, failed=0.646, completion_tokens=45.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 24%|██▍ | 97/400 [00:13<00:06, 44.77it/s, reward=-2.51, num_turns=1.18, num_tools=0.175, failed=0.649, completion_tokens=45.9]
validation: 24%|██▍ | 98/400 [00:13<00:06, 44.77it/s, reward=-2.51, num_turns=1.17, num_tools=0.173, failed=0.653, completion_tokens=45.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 25%|██▍ | 99/400 [00:13<00:06, 44.77it/s, reward=-2.52, num_turns=1.17, num_tools=0.172, failed=0.657, completion_tokens=45.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 25%|██▌ | 100/400 [00:13<00:06, 46.21it/s, reward=-2.52, num_turns=1.17, num_tools=0.172, failed=0.657, completion_tokens=45.9]
validation: 25%|██▌ | 100/400 [00:13<00:06, 46.21it/s, reward=-2.52, num_turns=1.17, num_tools=0.17, failed=0.66, completion_tokens=45.9]
validation: 25%|██▌ | 101/400 [00:13<00:06, 46.21it/s, reward=-2.53, num_turns=1.17, num_tools=0.168, failed=0.663, completion_tokens=45.9]
validation: 26%|██▌ | 102/400 [00:13<00:06, 46.21it/s, reward=-2.53, num_turns=1.17, num_tools=0.167, failed=0.667, completion_tokens=45.9]
validation: 26%|██▌ | 103/400 [00:13<00:06, 46.21it/s, reward=-2.54, num_turns=1.17, num_tools=0.165, failed=0.67, completion_tokens=45.9]
validation: 26%|██▌ | 104/400 [00:13<00:06, 46.21it/s, reward=-2.54, num_turns=1.16, num_tools=0.163, failed=0.673, completion_tokens=45.9]
validation: 26%|██▋ | 105/400 [00:13<00:06, 46.21it/s, reward=-2.54, num_turns=1.16, num_tools=0.162, failed=0.676, completion_tokens=45.9]
validation: 26%|██▋ | 106/400 [00:13<00:06, 46.21it/s, reward=-2.55, num_turns=1.16, num_tools=0.16, failed=0.679, completion_tokens=45.9]
validation: 27%|██▋ | 107/400 [00:13<00:06, 46.21it/s, reward=-2.54, num_turns=1.17, num_tools=0.168, failed=0.673, completion_tokens=45.3]
validation: 27%|██▋ | 108/400 [00:13<00:06, 46.21it/s, reward=-2.54, num_turns=1.18, num_tools=0.176, failed=0.667, completion_tokens=45]
validation: 27%|██▋ | 109/400 [00:13<00:06, 46.21it/s, reward=-2.53, num_turns=1.18, num_tools=0.183, failed=0.661, completion_tokens=44.7]
validation: 28%|██▊ | 110/400 [00:13<00:06, 46.21it/s, reward=-2.53, num_turns=1.19, num_tools=0.191, failed=0.655, completion_tokens=44.3][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 28%|██▊ | 111/400 [00:13<00:06, 46.21it/s, reward=-2.53, num_turns=1.19, num_tools=0.189, failed=0.658, completion_tokens=44.3]
validation: 28%|██▊ | 112/400 [00:13<00:06, 46.21it/s, reward=-2.54, num_turns=1.19, num_tools=0.188, failed=0.661, completion_tokens=44.3]
validation: 28%|██▊ | 113/400 [00:13<00:06, 46.21it/s, reward=-2.54, num_turns=1.19, num_tools=0.186, failed=0.664, completion_tokens=44.3]
validation: 28%|██▊ | 114/400 [00:13<00:06, 46.21it/s, reward=-2.55, num_turns=1.18, num_tools=0.184, failed=0.667, completion_tokens=44.3]
validation: 29%|██▉ | 115/400 [00:13<00:06, 46.21it/s, reward=-2.55, num_turns=1.18, num_tools=0.183, failed=0.67, completion_tokens=44.3]
validation: 29%|██▉ | 116/400 [00:13<00:06, 46.21it/s, reward=-2.55, num_turns=1.18, num_tools=0.181, failed=0.672, completion_tokens=44.3]
validation: 29%|██▉ | 117/400 [00:13<00:06, 46.21it/s, reward=-2.56, num_turns=1.18, num_tools=0.179, failed=0.675, completion_tokens=44.3]
validation: 30%|██▉ | 118/400 [00:13<00:06, 46.21it/s, reward=-2.56, num_turns=1.18, num_tools=0.178, failed=0.678, completion_tokens=44.3]
validation: 30%|██▉ | 119/400 [00:13<00:06, 46.21it/s, reward=-2.56, num_turns=1.18, num_tools=0.176, failed=0.681, completion_tokens=44.3]
validation: 30%|███ | 120/400 [00:13<00:06, 46.21it/s, reward=-2.57, num_turns=1.18, num_tools=0.175, failed=0.683, completion_tokens=44.3]
validation: 30%|███ | 121/400 [00:13<00:06, 46.21it/s, reward=-2.57, num_turns=1.17, num_tools=0.174, failed=0.686, completion_tokens=44.3]
validation: 30%|███ | 122/400 [00:13<00:06, 46.21it/s, reward=-2.58, num_turns=1.17, num_tools=0.172, failed=0.689, completion_tokens=44.3]
validation: 31%|███ | 123/400 [00:13<00:05, 46.21it/s, reward=-2.52, num_turns=1.18, num_tools=0.179, failed=0.683, completion_tokens=43.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 31%|███ | 124/400 [00:13<00:04, 67.91it/s, reward=-2.52, num_turns=1.18, num_tools=0.179, failed=0.683, completion_tokens=43.8]
validation: 31%|███ | 124/400 [00:13<00:04, 67.91it/s, reward=-2.52, num_turns=1.19, num_tools=0.185, failed=0.677, completion_tokens=44]
validation: 31%|███▏ | 125/400 [00:13<00:04, 67.91it/s, reward=-2.47, num_turns=1.19, num_tools=0.192, failed=0.672, completion_tokens=43.4]
validation: 32%|███▏ | 126/400 [00:13<00:04, 67.91it/s, reward=-2.46, num_turns=1.2, num_tools=0.198, failed=0.667, completion_tokens=43]
validation: 32%|███▏ | 127/400 [00:13<00:04, 67.91it/s, reward=-2.47, num_turns=1.2, num_tools=0.197, failed=0.669, completion_tokens=43][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 32%|███▏ | 128/400 [00:13<00:04, 67.91it/s, reward=-2.47, num_turns=1.2, num_tools=0.195, failed=0.672, completion_tokens=43]
validation: 32%|███▏ | 129/400 [00:13<00:03, 67.91it/s, reward=-2.47, num_turns=1.19, num_tools=0.194, failed=0.674, completion_tokens=43]
validation: 32%|███▎ | 130/400 [00:13<00:03, 67.91it/s, reward=-2.48, num_turns=1.2, num_tools=0.2, failed=0.677, completion_tokens=43]
validation: 33%|███▎ | 131/400 [00:13<00:03, 67.91it/s, reward=-2.48, num_turns=1.21, num_tools=0.206, failed=0.679, completion_tokens=42.8]
validation: 33%|███▎ | 132/400 [00:13<00:03, 67.91it/s, reward=-2.49, num_turns=1.21, num_tools=0.212, failed=0.682, completion_tokens=42.6]
validation: 33%|███▎ | 133/400 [00:13<00:03, 67.91it/s, reward=-2.49, num_turns=1.22, num_tools=0.218, failed=0.684, completion_tokens=42.4]
validation: 34%|███▎ | 134/400 [00:13<00:03, 67.91it/s, reward=-2.44, num_turns=1.22, num_tools=0.224, failed=0.679, completion_tokens=42.1]
validation: 34%|███▍ | 135/400 [00:13<00:03, 67.91it/s, reward=-2.45, num_turns=1.23, num_tools=0.23, failed=0.681, completion_tokens=42]
validation: 34%|███▍ | 136/400 [00:13<00:03, 67.91it/s, reward=-2.45, num_turns=1.24, num_tools=0.235, failed=0.684, completion_tokens=41.9]
validation: 34%|███▍ | 137/400 [00:13<00:03, 67.91it/s, reward=-2.45, num_turns=1.24, num_tools=0.241, failed=0.686, completion_tokens=41.9]
validation: 34%|███▍ | 138/400 [00:13<00:03, 67.91it/s, reward=-2.45, num_turns=1.25, num_tools=0.246, failed=0.681, completion_tokens=41.7]
validation: 35%|███▍ | 139/400 [00:13<00:03, 67.91it/s, reward=-2.4, num_turns=1.25, num_tools=0.252, failed=0.676, completion_tokens=41.4]
validation: 35%|███▌ | 140/400 [00:13<00:03, 67.91it/s, reward=-2.41, num_turns=1.25, num_tools=0.25, failed=0.671, completion_tokens=41.9]
validation: 35%|███▌ | 141/400 [00:13<00:03, 67.91it/s, reward=-2.41, num_turns=1.26, num_tools=0.255, failed=0.667, completion_tokens=41.8]
validation: 36%|███▌ | 142/400 [00:13<00:03, 67.91it/s, reward=-2.4, num_turns=1.26, num_tools=0.261, failed=0.662, completion_tokens=41.6]
validation: 36%|███▌ | 143/400 [00:13<00:03, 67.91it/s, reward=-2.36, num_turns=1.27, num_tools=0.266, failed=0.657, completion_tokens=41.3]
validation: 36%|███▌ | 144/400 [00:13<00:03, 67.91it/s, reward=-2.31, num_turns=1.27, num_tools=0.271, failed=0.653, completion_tokens=41.1]
validation: 36%|███▋ | 145/400 [00:13<00:03, 67.91it/s, reward=-2.27, num_turns=1.28, num_tools=0.276, failed=0.648, completion_tokens=40.9]
validation: 36%|███▋ | 146/400 [00:13<00:03, 67.91it/s, reward=-2.23, num_turns=1.28, num_tools=0.281, failed=0.644, completion_tokens=40.6]
validation: 37%|███▋ | 147/400 [00:13<00:03, 67.91it/s, reward=-2.23, num_turns=1.29, num_tools=0.286, failed=0.639, completion_tokens=40.4]
validation: 37%|███▋ | 148/400 [00:13<00:03, 67.91it/s, reward=-2.19, num_turns=1.29, num_tools=0.291, failed=0.635, completion_tokens=40.2]
validation: 37%|███▋ | 149/400 [00:13<00:03, 67.91it/s, reward=-2.16, num_turns=1.3, num_tools=0.295, failed=0.631, completion_tokens=39.9]
validation: 38%|███▊ | 150/400 [00:13<00:03, 67.91it/s, reward=-2.13, num_turns=1.3, num_tools=0.3, failed=0.627, completion_tokens=39.6]
validation: 38%|███▊ | 151/400 [00:13<00:03, 67.91it/s, reward=-2.1, num_turns=1.3, num_tools=0.305, failed=0.623, completion_tokens=39.4]
validation: 38%|███▊ | 152/400 [00:13<00:03, 67.91it/s, reward=-2.1, num_turns=1.31, num_tools=0.309, failed=0.618, completion_tokens=39.2]
validation: 38%|███▊ | 153/400 [00:13<00:03, 67.91it/s, reward=-2.1, num_turns=1.31, num_tools=0.314, failed=0.614, completion_tokens=39.1]
validation: 38%|███▊ | 154/400 [00:13<00:03, 67.91it/s, reward=-2.1, num_turns=1.32, num_tools=0.318, failed=0.61, completion_tokens=38.9]
validation: 39%|███▉ | 155/400 [00:13<00:03, 67.91it/s, reward=-2.08, num_turns=1.32, num_tools=0.323, failed=0.606, completion_tokens=38.8]
validation: 39%|███▉ | 156/400 [00:13<00:03, 67.91it/s, reward=-2.04, num_turns=1.33, num_tools=0.327, failed=0.603, completion_tokens=38.6]
validation: 39%|███▉ | 157/400 [00:13<00:03, 67.91it/s, reward=-2, num_turns=1.33, num_tools=0.331, failed=0.599, completion_tokens=38.5]
validation: 40%|███▉ | 158/400 [00:13<00:03, 67.91it/s, reward=-1.96, num_turns=1.34, num_tools=0.335, failed=0.595, completion_tokens=38.3]
validation: 40%|███▉ | 159/400 [00:13<00:03, 67.91it/s, reward=-1.92, num_turns=1.34, num_tools=0.34, failed=0.591, completion_tokens=38.2]
validation: 40%|████ | 160/400 [00:13<00:03, 67.91it/s, reward=-1.89, num_turns=1.34, num_tools=0.344, failed=0.588, completion_tokens=38.1]
validation: 40%|████ | 161/400 [00:13<00:03, 67.91it/s, reward=-1.85, num_turns=1.35, num_tools=0.348, failed=0.584, completion_tokens=38] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 40%|████ | 162/400 [00:13<00:02, 113.59it/s, reward=-1.85, num_turns=1.35, num_tools=0.348, failed=0.584, completion_tokens=38]
validation: 40%|████ | 162/400 [00:13<00:02, 113.59it/s, reward=-1.86, num_turns=1.35, num_tools=0.346, failed=0.586, completion_tokens=38]
validation: 41%|████ | 163/400 [00:13<00:02, 113.59it/s, reward=-1.87, num_turns=1.34, num_tools=0.344, failed=0.589, completion_tokens=38]
validation: 41%|████ | 164/400 [00:13<00:02, 113.59it/s, reward=-1.88, num_turns=1.34, num_tools=0.341, failed=0.591, completion_tokens=38]
validation: 41%|████▏ | 165/400 [00:13<00:02, 113.59it/s, reward=-1.88, num_turns=1.34, num_tools=0.339, failed=0.594, completion_tokens=38]
validation: 42%|████▏ | 166/400 [00:13<00:02, 113.59it/s, reward=-1.89, num_turns=1.34, num_tools=0.337, failed=0.596, completion_tokens=38]
validation: 42%|████▏ | 167/400 [00:13<00:02, 113.59it/s, reward=-1.9, num_turns=1.34, num_tools=0.335, failed=0.599, completion_tokens=38]
validation: 42%|████▏ | 168/400 [00:13<00:02, 113.59it/s, reward=-1.9, num_turns=1.33, num_tools=0.333, failed=0.601, completion_tokens=38]
validation: 42%|████▏ | 169/400 [00:13<00:02, 113.59it/s, reward=-1.91, num_turns=1.33, num_tools=0.331, failed=0.604, completion_tokens=38]
validation: 42%|████▎ | 170/400 [00:13<00:02, 113.59it/s, reward=-1.91, num_turns=1.34, num_tools=0.335, failed=0.6, completion_tokens=38.5]
validation: 43%|████▎ | 171/400 [00:13<00:02, 113.59it/s, reward=-1.92, num_turns=1.34, num_tools=0.339, failed=0.602, completion_tokens=38.2]
validation: 43%|████▎ | 172/400 [00:13<00:02, 113.59it/s, reward=-1.92, num_turns=1.34, num_tools=0.343, failed=0.605, completion_tokens=38]
validation: 43%|████▎ | 173/400 [00:13<00:01, 113.59it/s, reward=-1.93, num_turns=1.35, num_tools=0.347, failed=0.607, completion_tokens=37.8]
validation: 44%|████▎ | 174/400 [00:13<00:01, 113.59it/s, reward=-1.93, num_turns=1.35, num_tools=0.351, failed=0.609, completion_tokens=37.6]
validation: 44%|████▍ | 175/400 [00:13<00:01, 113.59it/s, reward=-1.94, num_turns=1.35, num_tools=0.354, failed=0.611, completion_tokens=37.4]
validation: 44%|████▍ | 176/400 [00:13<00:01, 113.59it/s, reward=-1.92, num_turns=1.36, num_tools=0.358, failed=0.608, completion_tokens=37.3]
validation: 44%|████▍ | 177/400 [00:13<00:01, 113.59it/s, reward=-1.9, num_turns=1.36, num_tools=0.362, failed=0.605, completion_tokens=37.1]
validation: 44%|████▍ | 178/400 [00:13<00:01, 113.59it/s, reward=-1.9, num_turns=1.37, num_tools=0.365, failed=0.607, completion_tokens=37]
validation: 45%|████▍ | 179/400 [00:13<00:01, 113.59it/s, reward=-1.9, num_turns=1.37, num_tools=0.369, failed=0.603, completion_tokens=36.9]
validation: 45%|████▌ | 180/400 [00:13<00:01, 113.59it/s, reward=-1.9, num_turns=1.37, num_tools=0.372, failed=0.6, completion_tokens=36.8]
validation: 45%|████▌ | 181/400 [00:14<00:01, 113.59it/s, reward=-1.9, num_turns=1.38, num_tools=0.376, failed=0.597, completion_tokens=37.6]
validation: 46%|████▌ | 182/400 [00:14<00:01, 113.59it/s, reward=-1.9, num_turns=1.38, num_tools=0.379, failed=0.593, completion_tokens=37.5][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 46%|████▌ | 183/400 [00:14<00:01, 113.59it/s, reward=-1.89, num_turns=1.38, num_tools=0.383, failed=0.59, completion_tokens=37.4]
validation: 46%|████▌ | 184/400 [00:14<00:01, 113.59it/s, reward=-1.88, num_turns=1.39, num_tools=0.386, failed=0.587, completion_tokens=37.3][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 46%|████▋ | 185/400 [00:14<00:01, 113.59it/s, reward=-1.88, num_turns=1.39, num_tools=0.389, failed=0.589, completion_tokens=37.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 46%|████▋ | 186/400 [00:14<00:01, 113.59it/s, reward=-1.89, num_turns=1.39, num_tools=0.392, failed=0.591, completion_tokens=37]
validation: 47%|████▋ | 187/400 [00:14<00:01, 113.59it/s, reward=-1.9, num_turns=1.4, num_tools=0.396, failed=0.594, completion_tokens=36.9]
validation: 47%|████▋ | 188/400 [00:14<00:01, 113.59it/s, reward=-1.9, num_turns=1.4, num_tools=0.399, failed=0.596, completion_tokens=36.8]
validation: 47%|████▋ | 189/400 [00:14<00:01, 113.59it/s, reward=-1.91, num_turns=1.4, num_tools=0.402, failed=0.598, completion_tokens=36.6]
validation: 48%|████▊ | 190/400 [00:14<00:01, 113.59it/s, reward=-1.91, num_turns=1.41, num_tools=0.405, failed=0.6, completion_tokens=36.5]
validation: 48%|████▊ | 191/400 [00:14<00:01, 113.59it/s, reward=-1.92, num_turns=1.41, num_tools=0.408, failed=0.602, completion_tokens=36.4]
validation: 48%|████▊ | 192/400 [00:14<00:01, 113.59it/s, reward=-1.92, num_turns=1.41, num_tools=0.411, failed=0.604, completion_tokens=36.2]
validation: 48%|████▊ | 193/400 [00:14<00:01, 113.59it/s, reward=-1.93, num_turns=1.41, num_tools=0.415, failed=0.606, completion_tokens=36.1]
validation: 48%|████▊ | 194/400 [00:14<00:01, 113.59it/s, reward=-1.9, num_turns=1.42, num_tools=0.418, failed=0.603, completion_tokens=36.1] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 49%|████▉ | 195/400 [00:14<00:01, 150.42it/s, reward=-1.9, num_turns=1.42, num_tools=0.418, failed=0.603, completion_tokens=36.1]
validation: 49%|████▉ | 195/400 [00:14<00:01, 150.42it/s, reward=-1.91, num_turns=1.42, num_tools=0.415, failed=0.605, completion_tokens=36.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 49%|████▉ | 196/400 [00:14<00:01, 150.42it/s, reward=-1.92, num_turns=1.42, num_tools=0.418, failed=0.607, completion_tokens=36] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 49%|████▉ | 197/400 [00:14<00:01, 150.42it/s, reward=-1.92, num_turns=1.42, num_tools=0.421, failed=0.609, completion_tokens=35.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 50%|████▉ | 198/400 [00:14<00:01, 150.42it/s, reward=-1.93, num_turns=1.42, num_tools=0.424, failed=0.611, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 50%|████▉ | 199/400 [00:14<00:01, 150.42it/s, reward=-1.93, num_turns=1.43, num_tools=0.427, failed=0.613, completion_tokens=35.6][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 50%|█████ | 200/400 [00:14<00:01, 150.42it/s, reward=-1.94, num_turns=1.43, num_tools=0.43, failed=0.615, completion_tokens=35.4] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 50%|█████ | 201/400 [00:14<00:01, 150.42it/s, reward=-1.94, num_turns=1.43, num_tools=0.433, failed=0.617, completion_tokens=35.3][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 50%|█████ | 202/400 [00:14<00:01, 150.42it/s, reward=-1.95, num_turns=1.44, num_tools=0.436, failed=0.619, completion_tokens=35.2]
validation: 51%|█████ | 203/400 [00:14<00:01, 150.42it/s, reward=-1.93, num_turns=1.44, num_tools=0.438, failed=0.616, completion_tokens=35.3]
validation: 51%|█████ | 204/400 [00:14<00:01, 150.42it/s, reward=-1.91, num_turns=1.44, num_tools=0.441, failed=0.613, completion_tokens=35.1]
validation: 51%|█████▏ | 205/400 [00:14<00:01, 150.42it/s, reward=-1.91, num_turns=1.44, num_tools=0.444, failed=0.61, completion_tokens=35.2]
validation: 52%|█████▏ | 206/400 [00:14<00:01, 150.42it/s, reward=-1.91, num_turns=1.45, num_tools=0.447, failed=0.607, completion_tokens=35]
validation: 52%|█████▏ | 207/400 [00:14<00:01, 150.42it/s, reward=-1.91, num_turns=1.45, num_tools=0.449, failed=0.604, completion_tokens=35.1]
validation: 52%|█████▏ | 208/400 [00:14<00:01, 150.42it/s, reward=-1.9, num_turns=1.45, num_tools=0.452, failed=0.601, completion_tokens=35]
validation: 52%|█████▏ | 209/400 [00:14<00:01, 150.42it/s, reward=-1.9, num_turns=1.45, num_tools=0.455, failed=0.598, completion_tokens=35]
validation: 52%|█████▎ | 210/400 [00:14<00:01, 150.42it/s, reward=-1.9, num_turns=1.46, num_tools=0.457, failed=0.595, completion_tokens=35]
validation: 53%|█████▎ | 211/400 [00:14<00:01, 150.42it/s, reward=-1.89, num_turns=1.46, num_tools=0.46, failed=0.592, completion_tokens=35]
validation: 53%|█████▎ | 212/400 [00:14<00:01, 150.42it/s, reward=-1.89, num_turns=1.46, num_tools=0.462, failed=0.59, completion_tokens=35]
validation: 53%|█████▎ | 213/400 [00:14<00:01, 150.42it/s, reward=-1.89, num_turns=1.46, num_tools=0.465, failed=0.587, completion_tokens=35][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 54%|█████▎ | 214/400 [00:14<00:01, 150.42it/s, reward=-1.89, num_turns=1.46, num_tools=0.463, failed=0.589, completion_tokens=35]
validation: 54%|█████▍ | 215/400 [00:14<00:01, 150.42it/s, reward=-1.9, num_turns=1.47, num_tools=0.465, failed=0.591, completion_tokens=34.9]
validation: 54%|█████▍ | 216/400 [00:14<00:01, 150.42it/s, reward=-1.9, num_turns=1.47, num_tools=0.468, failed=0.593, completion_tokens=34.8]
validation: 54%|█████▍ | 217/400 [00:14<00:01, 150.42it/s, reward=-1.91, num_turns=1.47, num_tools=0.47, failed=0.594, completion_tokens=34.7]
validation: 55%|█████▍ | 218/400 [00:14<00:01, 155.23it/s, reward=-1.91, num_turns=1.47, num_tools=0.47, failed=0.594, completion_tokens=34.7]
validation: 55%|█████▍ | 218/400 [00:14<00:01, 155.23it/s, reward=-1.91, num_turns=1.47, num_tools=0.472, failed=0.596, completion_tokens=34.6]
validation: 55%|█████▍ | 219/400 [00:14<00:01, 155.23it/s, reward=-1.92, num_turns=1.47, num_tools=0.475, failed=0.598, completion_tokens=34.6]
validation: 55%|█████▌ | 220/400 [00:14<00:01, 155.23it/s, reward=-1.92, num_turns=1.48, num_tools=0.477, failed=0.6, completion_tokens=34.5]
validation: 55%|█████▌ | 221/400 [00:14<00:01, 155.23it/s, reward=-1.93, num_turns=1.48, num_tools=0.48, failed=0.602, completion_tokens=34.4]
validation: 56%|█████▌ | 222/400 [00:14<00:01, 155.23it/s, reward=-1.93, num_turns=1.48, num_tools=0.482, failed=0.604, completion_tokens=34.3]
validation: 56%|█████▌ | 223/400 [00:14<00:01, 155.23it/s, reward=-1.94, num_turns=1.48, num_tools=0.48, failed=0.605, completion_tokens=34.3]
validation: 56%|█████▌ | 224/400 [00:14<00:01, 155.23it/s, reward=-1.94, num_turns=1.48, num_tools=0.482, failed=0.603, completion_tokens=34.3]
validation: 56%|█████▋ | 225/400 [00:14<00:01, 155.23it/s, reward=-1.94, num_turns=1.48, num_tools=0.484, failed=0.6, completion_tokens=34.3]
validation: 56%|█████▋ | 226/400 [00:14<00:01, 155.23it/s, reward=-1.92, num_turns=1.49, num_tools=0.487, failed=0.597, completion_tokens=34.2]
validation: 57%|█████▋ | 227/400 [00:14<00:01, 155.23it/s, reward=-1.91, num_turns=1.49, num_tools=0.489, failed=0.595, completion_tokens=34.3]
validation: 57%|█████▋ | 228/400 [00:14<00:01, 155.23it/s, reward=-1.88, num_turns=1.49, num_tools=0.491, failed=0.592, completion_tokens=34.3]
validation: 57%|█████▋ | 229/400 [00:14<00:01, 155.23it/s, reward=-1.86, num_turns=1.49, num_tools=0.493, failed=0.59, completion_tokens=34.2]
validation: 57%|█████▊ | 230/400 [00:14<00:01, 155.23it/s, reward=-1.83, num_turns=1.5, num_tools=0.496, failed=0.587, completion_tokens=34.2][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 58%|█████▊ | 231/400 [00:14<00:01, 155.23it/s, reward=-1.82, num_turns=1.5, num_tools=0.498, failed=0.584, completion_tokens=34.1]
validation: 58%|█████▊ | 232/400 [00:14<00:01, 155.23it/s, reward=-1.82, num_turns=1.5, num_tools=0.5, failed=0.586, completion_tokens=34.1]
validation: 58%|█████▊ | 233/400 [00:14<00:01, 155.23it/s, reward=-1.83, num_turns=1.5, num_tools=0.502, failed=0.588, completion_tokens=34]
validation: 58%|█████▊ | 234/400 [00:14<00:01, 155.23it/s, reward=-1.83, num_turns=1.5, num_tools=0.504, failed=0.59, completion_tokens=33.9]
validation: 59%|█████▉ | 235/400 [00:14<00:01, 155.23it/s, reward=-1.82, num_turns=1.51, num_tools=0.506, failed=0.587, completion_tokens=34.2]
validation: 59%|█████▉ | 236/400 [00:14<00:01, 155.23it/s, reward=-1.82, num_turns=1.51, num_tools=0.508, failed=0.585, completion_tokens=34.1]
validation: 59%|█████▉ | 237/400 [00:14<00:01, 155.23it/s, reward=-1.79, num_turns=1.51, num_tools=0.511, failed=0.582, completion_tokens=34.1]
validation: 60%|█████▉ | 238/400 [00:14<00:01, 155.23it/s, reward=-1.77, num_turns=1.51, num_tools=0.513, failed=0.58, completion_tokens=34]
validation: 60%|█████▉ | 239/400 [00:14<00:00, 163.47it/s, reward=-1.77, num_turns=1.51, num_tools=0.513, failed=0.58, completion_tokens=34]
validation: 60%|█████▉ | 239/400 [00:14<00:00, 163.47it/s, reward=-1.76, num_turns=1.51, num_tools=0.515, failed=0.577, completion_tokens=34.1]
validation: 60%|██████ | 240/400 [00:14<00:00, 163.47it/s, reward=-1.76, num_turns=1.52, num_tools=0.517, failed=0.575, completion_tokens=34.1]
validation: 60%|██████ | 241/400 [00:14<00:00, 163.47it/s, reward=-1.76, num_turns=1.52, num_tools=0.519, failed=0.573, completion_tokens=34]
validation: 60%|██████ | 242/400 [00:14<00:00, 163.47it/s, reward=-1.76, num_turns=1.52, num_tools=0.521, failed=0.57, completion_tokens=34]
validation: 61%|██████ | 243/400 [00:14<00:00, 163.47it/s, reward=-1.75, num_turns=1.52, num_tools=0.523, failed=0.568, completion_tokens=34.1]
validation: 61%|██████ | 244/400 [00:14<00:00, 163.47it/s, reward=-1.75, num_turns=1.52, num_tools=0.525, failed=0.566, completion_tokens=34.2]
validation: 61%|██████▏ | 245/400 [00:14<00:00, 163.47it/s, reward=-1.75, num_turns=1.53, num_tools=0.527, failed=0.563, completion_tokens=34.1]
validation: 62%|██████▏ | 246/400 [00:14<00:00, 163.47it/s, reward=-1.75, num_turns=1.53, num_tools=0.528, failed=0.561, completion_tokens=34.2]
validation: 62%|██████▏ | 247/400 [00:14<00:00, 163.47it/s, reward=-1.75, num_turns=1.53, num_tools=0.53, failed=0.559, completion_tokens=34.2] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 62%|██████▏ | 248/400 [00:14<00:00, 163.47it/s, reward=-1.76, num_turns=1.53, num_tools=0.532, failed=0.56, completion_tokens=34.3]
validation: 62%|██████▏ | 249/400 [00:50<00:00, 163.47it/s, reward=-1.76, num_turns=1.53, num_tools=0.534, failed=0.562, completion_tokens=34.3]
validation: 62%|██████▏ | 249/400 [00:50<00:00, 163.47it/s, reward=-1.76, num_turns=1.53, num_tools=0.534, failed=0.562, completion_tokens=34.3]
validation: 62%|██████▎ | 250/400 [00:50<01:21, 1.85it/s, reward=-1.76, num_turns=1.53, num_tools=0.534, failed=0.562, completion_tokens=34.3]
validation: 62%|██████▎ | 250/400 [00:50<01:21, 1.85it/s, reward=-1.77, num_turns=1.54, num_tools=0.536, failed=0.564, completion_tokens=34.2]
validation: 63%|██████▎ | 251/400 [00:50<01:19, 1.88it/s, reward=-1.77, num_turns=1.54, num_tools=0.536, failed=0.564, completion_tokens=34.2]
validation: 63%|██████▎ | 251/400 [00:50<01:19, 1.88it/s, reward=-1.77, num_turns=1.54, num_tools=0.538, failed=0.562, completion_tokens=34.2]
validation: 63%|██████▎ | 252/400 [00:50<01:18, 1.88it/s, reward=-1.77, num_turns=1.54, num_tools=0.54, failed=0.56, completion_tokens=34.2] <string>:1: SyntaxWarning: 'float' object is not callable; perhaps you missed a comma?
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 63%|██████▎ | 253/400 [00:50<01:18, 1.88it/s, reward=-1.77, num_turns=1.54, num_tools=0.542, failed=0.557, completion_tokens=34.4]
validation: 64%|██████▎ | 254/400 [00:50<01:17, 1.88it/s, reward=-1.77, num_turns=1.54, num_tools=0.543, failed=0.555, completion_tokens=34.5]
validation: 64%|██████▍ | 255/400 [00:50<01:17, 1.88it/s, reward=-1.77, num_turns=1.55, num_tools=0.545, failed=0.553, completion_tokens=34.7]
validation: 64%|██████▍ | 256/400 [00:50<01:16, 1.88it/s, reward=-1.78, num_turns=1.55, num_tools=0.547, failed=0.555, completion_tokens=34.6]
validation: 64%|██████▍ | 257/400 [00:50<01:16, 1.88it/s, reward=-1.78, num_turns=1.55, num_tools=0.549, failed=0.553, completion_tokens=34.7]
validation: 64%|██████▍ | 258/400 [00:50<01:15, 1.88it/s, reward=-1.78, num_turns=1.55, num_tools=0.55, failed=0.55, completion_tokens=34.8]
validation: 65%|██████▍ | 259/400 [00:50<01:15, 1.88it/s, reward=-1.78, num_turns=1.55, num_tools=0.552, failed=0.548, completion_tokens=35]
validation: 65%|██████▌ | 260/400 [00:50<01:14, 1.88it/s, reward=-1.78, num_turns=1.55, num_tools=0.554, failed=0.546, completion_tokens=35.1]
validation: 65%|██████▌ | 261/400 [00:50<01:14, 1.88it/s, reward=-1.78, num_turns=1.56, num_tools=0.556, failed=0.544, completion_tokens=35.1]
validation: 66%|██████▌ | 262/400 [00:50<01:13, 1.88it/s, reward=-1.78, num_turns=1.56, num_tools=0.557, failed=0.542, completion_tokens=35.1]
validation: 66%|██████▌ | 263/400 [00:50<01:13, 1.88it/s, reward=-1.77, num_turns=1.56, num_tools=0.559, failed=0.54, completion_tokens=35.3]
validation: 66%|██████▌ | 264/400 [00:50<01:12, 1.88it/s, reward=-1.76, num_turns=1.56, num_tools=0.561, failed=0.538, completion_tokens=35.3]
validation: 66%|██████▋ | 265/400 [00:50<01:11, 1.88it/s, reward=-1.75, num_turns=1.56, num_tools=0.562, failed=0.536, completion_tokens=35.3]
validation: 66%|██████▋ | 266/400 [00:50<00:51, 2.62it/s, reward=-1.75, num_turns=1.56, num_tools=0.562, failed=0.536, completion_tokens=35.3]
validation: 66%|██████▋ | 266/400 [00:50<00:51, 2.62it/s, reward=-1.73, num_turns=1.56, num_tools=0.564, failed=0.534, completion_tokens=35.3]
validation: 67%|██████▋ | 267/400 [00:50<00:50, 2.62it/s, reward=-1.72, num_turns=1.57, num_tools=0.566, failed=0.532, completion_tokens=35.3]
validation: 67%|██████▋ | 268/400 [00:50<00:50, 2.62it/s, reward=-1.71, num_turns=1.57, num_tools=0.567, failed=0.53, completion_tokens=35.4]
validation: 67%|██████▋ | 269/400 [00:50<00:49, 2.62it/s, reward=-1.7, num_turns=1.57, num_tools=0.569, failed=0.528, completion_tokens=35.7]
validation: 68%|██████▊ | 270/400 [00:50<00:49, 2.62it/s, reward=-1.69, num_turns=1.57, num_tools=0.57, failed=0.526, completion_tokens=36.2]
validation: 68%|██████▊ | 271/400 [00:50<00:49, 2.62it/s, reward=-1.68, num_turns=1.57, num_tools=0.572, failed=0.524, completion_tokens=36.3]
validation: 68%|██████▊ | 272/400 [00:50<00:48, 2.62it/s, reward=-1.68, num_turns=1.57, num_tools=0.574, failed=0.522, completion_tokens=36.3]
validation: 68%|██████▊ | 273/400 [00:50<00:48, 2.62it/s, reward=-1.66, num_turns=1.58, num_tools=0.575, failed=0.52, completion_tokens=36.4]
validation: 68%|██████▊ | 274/400 [00:50<00:48, 2.62it/s, reward=-1.65, num_turns=1.58, num_tools=0.577, failed=0.518, completion_tokens=36.7]
validation: 69%|██████▉ | 275/400 [00:50<00:47, 2.62it/s, reward=-1.63, num_turns=1.58, num_tools=0.582, failed=0.516, completion_tokens=36.8]
validation: 69%|██████▉ | 276/400 [00:50<00:47, 2.62it/s, reward=-1.62, num_turns=1.58, num_tools=0.583, failed=0.514, completion_tokens=36.8]
validation: 69%|██████▉ | 277/400 [00:50<00:46, 2.62it/s, reward=-1.61, num_turns=1.58, num_tools=0.585, failed=0.513, completion_tokens=36.9]
validation: 70%|██████▉ | 278/400 [00:50<00:46, 2.62it/s, reward=-1.61, num_turns=1.59, num_tools=0.586, failed=0.511, completion_tokens=37.1]
validation: 70%|██████▉ | 279/400 [00:50<00:46, 2.62it/s, reward=-1.61, num_turns=1.59, num_tools=0.588, failed=0.509, completion_tokens=37.4]
validation: 70%|███████ | 280/400 [00:50<00:45, 2.62it/s, reward=-1.61, num_turns=1.59, num_tools=0.589, failed=0.507, completion_tokens=37.7]
validation: 70%|███████ | 281/400 [00:51<00:32, 3.71it/s, reward=-1.61, num_turns=1.59, num_tools=0.589, failed=0.507, completion_tokens=37.7]
validation: 70%|███████ | 281/400 [00:51<00:32, 3.71it/s, reward=-1.62, num_turns=1.59, num_tools=0.591, failed=0.505, completion_tokens=37.7]
validation: 70%|███████ | 282/400 [00:51<00:31, 3.71it/s, reward=-1.62, num_turns=1.59, num_tools=0.592, failed=0.504, completion_tokens=37.7]
validation: 71%|███████ | 283/400 [00:51<00:31, 3.71it/s, reward=-1.62, num_turns=1.59, num_tools=0.594, failed=0.502, completion_tokens=37.7]
validation: 71%|███████ | 284/400 [00:51<00:31, 3.71it/s, reward=-1.62, num_turns=1.6, num_tools=0.595, failed=0.5, completion_tokens=37.8]
validation: 71%|███████▏ | 285/400 [00:51<00:31, 3.71it/s, reward=-1.61, num_turns=1.6, num_tools=0.596, failed=0.498, completion_tokens=37.8]
validation: 72%|███████▏ | 286/400 [00:51<00:30, 3.71it/s, reward=-1.59, num_turns=1.6, num_tools=0.598, failed=0.497, completion_tokens=37.7]
validation: 72%|███████▏ | 287/400 [00:51<00:30, 3.71it/s, reward=-1.57, num_turns=1.6, num_tools=0.599, failed=0.495, completion_tokens=37.7]
validation: 72%|███████▏ | 288/400 [00:51<00:30, 3.71it/s, reward=-1.55, num_turns=1.6, num_tools=0.601, failed=0.493, completion_tokens=37.6]
validation: 72%|███████▏ | 289/400 [00:51<00:29, 3.71it/s, reward=-1.54, num_turns=1.6, num_tools=0.602, failed=0.491, completion_tokens=37.6]
validation: 72%|███████▎ | 290/400 [00:51<00:29, 3.71it/s, reward=-1.53, num_turns=1.6, num_tools=0.603, failed=0.49, completion_tokens=37.8]
validation: 73%|███████▎ | 291/400 [00:51<00:22, 4.74it/s, reward=-1.53, num_turns=1.6, num_tools=0.603, failed=0.49, completion_tokens=37.8]
validation: 73%|███████▎ | 291/400 [00:51<00:22, 4.74it/s, reward=-1.53, num_turns=1.6, num_tools=0.605, failed=0.488, completion_tokens=37.8]
validation: 73%|███████▎ | 292/400 [00:51<00:22, 4.74it/s, reward=-1.52, num_turns=1.61, num_tools=0.606, failed=0.486, completion_tokens=38]
validation: 73%|███████▎ | 293/400 [00:51<00:22, 4.74it/s, reward=-1.52, num_turns=1.61, num_tools=0.608, failed=0.485, completion_tokens=38]
validation: 74%|███████▎ | 294/400 [00:51<00:22, 4.74it/s, reward=-1.51, num_turns=1.61, num_tools=0.609, failed=0.483, completion_tokens=37.9]
validation: 74%|███████▍ | 295/400 [00:51<00:22, 4.74it/s, reward=-1.5, num_turns=1.61, num_tools=0.61, failed=0.481, completion_tokens=37.8]
validation: 74%|███████▍ | 296/400 [00:51<00:21, 4.74it/s, reward=-1.49, num_turns=1.61, num_tools=0.611, failed=0.48, completion_tokens=37.8]
validation: 74%|███████▍ | 297/400 [00:51<00:21, 4.74it/s, reward=-1.49, num_turns=1.61, num_tools=0.613, failed=0.478, completion_tokens=37.7]
validation: 74%|███████▍ | 298/400 [00:51<00:21, 4.74it/s, reward=-1.49, num_turns=1.61, num_tools=0.614, failed=0.477, completion_tokens=37.7]
validation: 75%|███████▍ | 299/400 [00:51<00:21, 4.74it/s, reward=-1.49, num_turns=1.62, num_tools=0.615, failed=0.475, completion_tokens=37.7]
validation: 75%|███████▌ | 300/400 [00:51<00:21, 4.74it/s, reward=-1.48, num_turns=1.62, num_tools=0.617, failed=0.473, completion_tokens=37.7]
validation: 75%|███████▌ | 301/400 [00:51<00:20, 4.74it/s, reward=-1.48, num_turns=1.62, num_tools=0.618, failed=0.472, completion_tokens=38]
validation: 76%|███████▌ | 302/400 [00:51<00:20, 4.74it/s, reward=-1.47, num_turns=1.62, num_tools=0.619, failed=0.47, completion_tokens=38]
validation: 76%|███████▌ | 303/400 [00:51<00:20, 4.74it/s, reward=-1.45, num_turns=1.62, num_tools=0.62, failed=0.469, completion_tokens=37.9]
validation: 76%|███████▌ | 304/400 [00:51<00:20, 4.74it/s, reward=-1.43, num_turns=1.62, num_tools=0.622, failed=0.467, completion_tokens=37.9]
validation: 76%|███████▋ | 305/400 [00:51<00:20, 4.74it/s, reward=-1.41, num_turns=1.62, num_tools=0.623, failed=0.466, completion_tokens=37.8]
validation: 76%|███████▋ | 306/400 [00:51<00:19, 4.74it/s, reward=-1.4, num_turns=1.62, num_tools=0.624, failed=0.464, completion_tokens=37.8]
validation: 77%|███████▋ | 307/400 [00:51<00:19, 4.74it/s, reward=-1.4, num_turns=1.63, num_tools=0.625, failed=0.463, completion_tokens=37.8]
validation: 77%|███████▋ | 308/400 [00:51<00:19, 4.74it/s, reward=-1.4, num_turns=1.63, num_tools=0.627, failed=0.461, completion_tokens=37.8]
validation: 77%|███████▋ | 309/400 [00:51<00:19, 4.74it/s, reward=-1.4, num_turns=1.63, num_tools=0.628, failed=0.46, completion_tokens=37.8]
validation: 78%|███████▊ | 310/400 [00:51<00:18, 4.74it/s, reward=-1.41, num_turns=1.63, num_tools=0.626, failed=0.458, completion_tokens=38.2]
validation: 78%|███████▊ | 311/400 [00:51<00:18, 4.74it/s, reward=-1.41, num_turns=1.63, num_tools=0.627, failed=0.457, completion_tokens=38.3]
validation: 78%|███████▊ | 312/400 [00:51<00:18, 4.74it/s, reward=-1.41, num_turns=1.63, num_tools=0.628, failed=0.455, completion_tokens=38.4]
validation: 78%|███████▊ | 313/400 [00:51<00:18, 4.74it/s, reward=-1.41, num_turns=1.63, num_tools=0.629, failed=0.454, completion_tokens=38.8]
validation: 78%|███████▊ | 314/400 [00:51<00:18, 4.74it/s, reward=-1.42, num_turns=1.63, num_tools=0.627, failed=0.452, completion_tokens=39.1]
validation: 79%|███████▉ | 315/400 [00:51<00:17, 4.74it/s, reward=-1.42, num_turns=1.63, num_tools=0.629, failed=0.451, completion_tokens=39.1]
validation: 79%|███████▉ | 316/400 [00:51<00:17, 4.74it/s, reward=-1.42, num_turns=1.63, num_tools=0.63, failed=0.449, completion_tokens=39.1]
validation: 79%|███████▉ | 317/400 [00:51<00:17, 4.74it/s, reward=-1.42, num_turns=1.63, num_tools=0.631, failed=0.448, completion_tokens=39.1]
validation: 80%|███████▉ | 318/400 [00:51<00:17, 4.74it/s, reward=-1.43, num_turns=1.63, num_tools=0.632, failed=0.447, completion_tokens=39.2]
validation: 80%|███████▉ | 319/400 [00:51<00:17, 4.74it/s, reward=-1.43, num_turns=1.63, num_tools=0.633, failed=0.445, completion_tokens=39.2]
validation: 80%|████████ | 320/400 [00:51<00:16, 4.74it/s, reward=-1.43, num_turns=1.63, num_tools=0.634, failed=0.444, completion_tokens=39.2]
validation: 80%|████████ | 321/400 [00:51<00:16, 4.74it/s, reward=-1.42, num_turns=1.64, num_tools=0.636, failed=0.442, completion_tokens=39.1]
validation: 80%|████████ | 322/400 [00:51<00:16, 4.74it/s, reward=-1.41, num_turns=1.64, num_tools=0.637, failed=0.441, completion_tokens=39.1]
validation: 81%|████████ | 323/400 [00:51<00:16, 4.74it/s, reward=-1.4, num_turns=1.64, num_tools=0.638, failed=0.44, completion_tokens=39]
validation: 81%|████████ | 324/400 [00:51<00:16, 4.74it/s, reward=-1.4, num_turns=1.64, num_tools=0.639, failed=0.438, completion_tokens=38.9]
validation: 81%|████████▏ | 325/400 [00:51<00:15, 4.74it/s, reward=-1.4, num_turns=1.64, num_tools=0.64, failed=0.437, completion_tokens=38.9]
validation: 82%|████████▏ | 326/400 [00:51<00:15, 4.74it/s, reward=-1.4, num_turns=1.64, num_tools=0.641, failed=0.436, completion_tokens=38.9]
validation: 82%|████████▏ | 327/400 [00:51<00:15, 4.74it/s, reward=-1.41, num_turns=1.64, num_tools=0.642, failed=0.434, completion_tokens=38.9]
validation: 82%|████████▏ | 328/400 [00:51<00:15, 4.74it/s, reward=-1.41, num_turns=1.64, num_tools=0.643, failed=0.433, completion_tokens=39]
validation: 82%|████████▏ | 329/400 [00:51<00:14, 4.74it/s, reward=-1.41, num_turns=1.64, num_tools=0.644, failed=0.432, completion_tokens=39]
validation: 82%|████████▎ | 330/400 [00:51<00:14, 4.74it/s, reward=-1.41, num_turns=1.65, num_tools=0.645, failed=0.43, completion_tokens=39]
validation: 83%|████████▎ | 331/400 [00:51<00:14, 4.74it/s, reward=-1.41, num_turns=1.65, num_tools=0.647, failed=0.429, completion_tokens=39]
validation: 83%|████████▎ | 332/400 [00:51<00:14, 4.74it/s, reward=-1.41, num_turns=1.65, num_tools=0.648, failed=0.428, completion_tokens=38.9]
validation: 83%|████████▎ | 333/400 [00:51<00:14, 4.74it/s, reward=-1.4, num_turns=1.65, num_tools=0.649, failed=0.426, completion_tokens=38.9]
validation: 84%|████████▎ | 334/400 [00:51<00:13, 4.74it/s, reward=-1.41, num_turns=1.65, num_tools=0.647, failed=0.425, completion_tokens=39]
validation: 84%|████████▍ | 335/400 [00:51<00:13, 4.74it/s, reward=-1.4, num_turns=1.65, num_tools=0.648, failed=0.424, completion_tokens=39.4]
validation: 84%|████████▍ | 336/400 [00:51<00:13, 4.74it/s, reward=-1.39, num_turns=1.65, num_tools=0.649, failed=0.423, completion_tokens=39.4]
validation: 84%|████████▍ | 337/400 [00:51<00:13, 4.74it/s, reward=-1.39, num_turns=1.65, num_tools=0.65, failed=0.421, completion_tokens=39.4]
validation: 84%|████████▍ | 338/400 [00:51<00:13, 4.74it/s, reward=-1.39, num_turns=1.65, num_tools=0.651, failed=0.42, completion_tokens=39.4]
validation: 85%|████████▍ | 339/400 [00:51<00:12, 4.74it/s, reward=-1.39, num_turns=1.65, num_tools=0.652, failed=0.419, completion_tokens=39.4]
validation: 85%|████████▌ | 340/400 [00:51<00:12, 4.74it/s, reward=-1.39, num_turns=1.65, num_tools=0.65, failed=0.418, completion_tokens=40.3]
validation: 85%|████████▌ | 341/400 [00:51<00:12, 4.74it/s, reward=-1.4, num_turns=1.65, num_tools=0.651, failed=0.416, completion_tokens=40.4]
validation: 86%|████████▌ | 342/400 [00:51<00:12, 4.74it/s, reward=-1.4, num_turns=1.65, num_tools=0.652, failed=0.415, completion_tokens=40.5]
validation: 86%|████████▌ | 343/400 [00:51<00:12, 4.74it/s, reward=-1.39, num_turns=1.65, num_tools=0.653, failed=0.414, completion_tokens=40.6]
validation: 86%|████████▌ | 344/400 [00:51<00:11, 4.74it/s, reward=-1.39, num_turns=1.65, num_tools=0.654, failed=0.413, completion_tokens=41]
validation: 86%|████████▋ | 345/400 [00:51<00:11, 4.74it/s, reward=-1.39, num_turns=1.66, num_tools=0.655, failed=0.412, completion_tokens=41.1]
validation: 86%|████████▋ | 346/400 [00:51<00:11, 4.74it/s, reward=-1.39, num_turns=1.66, num_tools=0.656, failed=0.41, completion_tokens=41.1]
validation: 87%|████████▋ | 347/400 [00:51<00:11, 4.74it/s, reward=-1.4, num_turns=1.65, num_tools=0.654, failed=0.409, completion_tokens=41.8]
validation: 87%|████████▋ | 348/400 [00:51<00:10, 4.74it/s, reward=-1.39, num_turns=1.66, num_tools=0.655, failed=0.408, completion_tokens=42.1]
validation: 87%|████████▋ | 349/400 [00:51<00:10, 4.74it/s, reward=-1.39, num_turns=1.66, num_tools=0.656, failed=0.407, completion_tokens=43.2]
validation: 88%|████████▊ | 350/400 [00:51<00:10, 4.74it/s, reward=-1.4, num_turns=1.65, num_tools=0.654, failed=0.406, completion_tokens=44.4]
validation: 88%|████████▊ | 351/400 [00:51<00:10, 4.74it/s, reward=-1.4, num_turns=1.65, num_tools=0.652, failed=0.405, completion_tokens=45.4]
validation: 88%|████████▊ | 352/400 [00:51<00:10, 4.74it/s, reward=-1.41, num_turns=1.65, num_tools=0.651, failed=0.403, completion_tokens=47.3]
validation: 88%|████████▊ | 353/400 [00:51<00:09, 4.74it/s, reward=-1.41, num_turns=1.65, num_tools=0.652, failed=0.402, completion_tokens=47.8]
validation: 88%|████████▊ | 354/400 [00:51<00:09, 4.74it/s, reward=-1.41, num_turns=1.65, num_tools=0.65, failed=0.401, completion_tokens=49.6]
validation: 89%|████████▉ | 355/400 [00:51<00:09, 4.74it/s, reward=-1.42, num_turns=1.65, num_tools=0.648, failed=0.4, completion_tokens=51.4]
validation: 89%|████████▉ | 356/400 [00:51<00:09, 4.74it/s, reward=-1.42, num_turns=1.65, num_tools=0.646, failed=0.399, completion_tokens=53.2]
validation: 89%|████████▉ | 357/400 [00:51<00:09, 4.74it/s, reward=-1.43, num_turns=1.64, num_tools=0.644, failed=0.398, completion_tokens=54.4]
validation: 90%|████████▉ | 358/400 [00:51<00:08, 4.74it/s, reward=-1.43, num_turns=1.65, num_tools=0.645, failed=0.397, completion_tokens=54.7]
validation: 90%|████████▉ | 359/400 [00:51<00:08, 4.74it/s, reward=-1.42, num_turns=1.65, num_tools=0.646, failed=0.396, completion_tokens=55.1]
validation: 90%|█████████ | 360/400 [00:51<00:08, 4.74it/s, reward=-1.43, num_turns=1.64, num_tools=0.644, failed=0.394, completion_tokens=56.4]
validation: 90%|█████████ | 361/400 [00:51<00:08, 4.74it/s, reward=-1.43, num_turns=1.64, num_tools=0.643, failed=0.393, completion_tokens=58.2]
validation: 90%|█████████ | 362/400 [00:51<00:08, 4.74it/s, reward=-1.44, num_turns=1.64, num_tools=0.641, failed=0.392, completion_tokens=59.9]
validation: 91%|█████████ | 363/400 [00:51<00:02, 14.97it/s, reward=-1.44, num_turns=1.64, num_tools=0.641, failed=0.392, completion_tokens=59.9]
validation: 91%|█████████ | 363/400 [00:51<00:02, 14.97it/s, reward=-1.44, num_turns=1.64, num_tools=0.642, failed=0.391, completion_tokens=59.8]
validation: 91%|█████████ | 364/400 [00:51<00:02, 14.97it/s, reward=-1.43, num_turns=1.64, num_tools=0.643, failed=0.39, completion_tokens=59.6]
validation: 91%|█████████▏| 365/400 [00:51<00:02, 14.97it/s, reward=-1.42, num_turns=1.64, num_tools=0.644, failed=0.389, completion_tokens=59.5]
validation: 92%|█████████▏| 366/400 [00:51<00:02, 14.97it/s, reward=-1.42, num_turns=1.64, num_tools=0.645, failed=0.388, completion_tokens=59.4]
validation: 92%|█████████▏| 367/400 [00:51<00:02, 14.97it/s, reward=-1.41, num_turns=1.65, num_tools=0.646, failed=0.387, completion_tokens=59.3]
validation: 92%|█████████▏| 368/400 [00:51<00:02, 14.97it/s, reward=-1.4, num_turns=1.65, num_tools=0.647, failed=0.386, completion_tokens=59.1]
validation: 92%|█████████▏| 369/400 [00:51<00:02, 14.97it/s, reward=-1.39, num_turns=1.65, num_tools=0.648, failed=0.385, completion_tokens=59]
validation: 92%|█████████▎| 370/400 [00:51<00:02, 14.97it/s, reward=-1.4, num_turns=1.65, num_tools=0.649, failed=0.384, completion_tokens=58.9]
validation: 93%|█████████▎| 371/400 [00:51<00:01, 14.97it/s, reward=-1.38, num_turns=1.65, num_tools=0.65, failed=0.383, completion_tokens=58.7]
validation: 93%|█████████▎| 372/400 [00:51<00:01, 14.97it/s, reward=-1.37, num_turns=1.65, num_tools=0.651, failed=0.382, completion_tokens=58.6]
validation: 93%|█████████▎| 373/400 [00:51<00:01, 14.97it/s, reward=-1.36, num_turns=1.65, num_tools=0.651, failed=0.381, completion_tokens=58.5]
validation: 94%|█████████▎| 374/400 [00:51<00:01, 14.97it/s, reward=-1.35, num_turns=1.65, num_tools=0.652, failed=0.38, completion_tokens=58.3]
validation: 94%|█████████▍| 375/400 [00:51<00:01, 14.97it/s, reward=-1.34, num_turns=1.65, num_tools=0.653, failed=0.379, completion_tokens=58.2]
validation: 94%|█████████▍| 376/400 [00:51<00:01, 14.97it/s, reward=-1.33, num_turns=1.65, num_tools=0.654, failed=0.378, completion_tokens=58.1]
validation: 94%|█████████▍| 377/400 [00:51<00:01, 14.97it/s, reward=-1.33, num_turns=1.66, num_tools=0.655, failed=0.377, completion_tokens=58]
validation: 94%|█████████▍| 378/400 [00:51<00:01, 14.97it/s, reward=-1.31, num_turns=1.66, num_tools=0.656, failed=0.376, completion_tokens=57.9]
validation: 95%|█████████▍| 379/400 [00:51<00:01, 14.97it/s, reward=-1.32, num_turns=1.66, num_tools=0.657, failed=0.375, completion_tokens=57.8]
validation: 95%|█████████▌| 380/400 [00:51<00:01, 14.97it/s, reward=-1.3, num_turns=1.66, num_tools=0.658, failed=0.374, completion_tokens=57.7]
validation: 95%|█████████▌| 381/400 [00:51<00:01, 14.97it/s, reward=-1.31, num_turns=1.66, num_tools=0.659, failed=0.373, completion_tokens=57.6]
validation: 96%|█████████▌| 382/400 [00:51<00:01, 14.97it/s, reward=-1.31, num_turns=1.66, num_tools=0.66, failed=0.372, completion_tokens=57.5]
validation: 96%|█████████▌| 383/400 [00:51<00:01, 14.97it/s, reward=-1.31, num_turns=1.66, num_tools=0.661, failed=0.371, completion_tokens=57.4]
validation: 96%|█████████▌| 384/400 [00:51<00:01, 14.97it/s, reward=-1.3, num_turns=1.66, num_tools=0.661, failed=0.37, completion_tokens=57.2]
validation: 96%|█████████▋| 385/400 [00:51<00:01, 14.97it/s, reward=-1.3, num_turns=1.66, num_tools=0.662, failed=0.369, completion_tokens=57.1]
validation: 96%|█████████▋| 386/400 [00:51<00:00, 19.34it/s, reward=-1.3, num_turns=1.66, num_tools=0.662, failed=0.369, completion_tokens=57.1]
validation: 96%|█████████▋| 386/400 [00:51<00:00, 19.34it/s, reward=-1.3, num_turns=1.66, num_tools=0.663, failed=0.368, completion_tokens=57.1]
validation: 97%|█████████▋| 387/400 [00:51<00:00, 19.34it/s, reward=-1.3, num_turns=1.66, num_tools=0.664, failed=0.367, completion_tokens=57]
validation: 97%|█████████▋| 388/400 [00:51<00:00, 19.34it/s, reward=-1.3, num_turns=1.66, num_tools=0.665, failed=0.366, completion_tokens=56.9]
validation: 97%|█████████▋| 389/400 [00:51<00:00, 19.34it/s, reward=-1.29, num_turns=1.67, num_tools=0.666, failed=0.365, completion_tokens=56.8]
validation: 98%|█████████▊| 390/400 [00:51<00:00, 19.34it/s, reward=-1.29, num_turns=1.67, num_tools=0.667, failed=0.364, completion_tokens=56.8]
validation: 98%|█████████▊| 391/400 [00:51<00:00, 19.34it/s, reward=-1.29, num_turns=1.67, num_tools=0.668, failed=0.363, completion_tokens=56.7]
validation: 98%|█████████▊| 392/400 [00:51<00:00, 19.34it/s, reward=-1.3, num_turns=1.67, num_tools=0.668, failed=0.362, completion_tokens=56.7]
validation: 98%|█████████▊| 393/400 [00:51<00:00, 19.34it/s, reward=-1.29, num_turns=1.67, num_tools=0.672, failed=0.361, completion_tokens=56.8]
validation: 98%|█████████▊| 394/400 [00:51<00:00, 19.34it/s, reward=-1.29, num_turns=1.67, num_tools=0.673, failed=0.36, completion_tokens=56.8]
validation: 99%|█████████▉| 395/400 [00:52<00:00, 19.34it/s, reward=-1.29, num_turns=1.67, num_tools=0.673, failed=0.359, completion_tokens=56.8]
validation: 99%|█████████▉| 396/400 [00:52<00:00, 19.34it/s, reward=-1.3, num_turns=1.67, num_tools=0.674, failed=0.359, completion_tokens=56.9]
validation: 99%|█████████▉| 397/400 [00:52<00:00, 19.34it/s, reward=-1.3, num_turns=1.67, num_tools=0.675, failed=0.358, completion_tokens=57.1]
validation: 100%|█████████▉| 398/400 [00:52<00:00, 19.34it/s, reward=-1.29, num_turns=1.67, num_tools=0.676, failed=0.357, completion_tokens=57.2]
validation: 100%|█████████▉| 399/400 [00:52<00:00, 19.34it/s, reward=-1.29, num_turns=1.67, num_tools=0.677, failed=0.356, completion_tokens=57.7]
validation: 100%|██████████| 400/400 [00:52<00:00, 19.34it/s, reward=-1.3, num_turns=1.68, num_tools=0.677, failed=0.355, completion_tokens=58.1]
validation: 100%|██████████| 400/400 [00:52<00:00, 7.55it/s, reward=-1.3, num_turns=1.68, num_tools=0.677, failed=0.355, completion_tokens=58.1]
Val avg reward: -1.295
============================================================
Step 6/50
============================================================
step 6: 0%| | 0/32 [00:00<?, ?it/s]
step 6: 3%|▎ | 1/32 [00:01<00:40, 1.29s/it]
step 6: 3%|▎ | 1/32 [00:01<00:40, 1.29s/it, reward=2.5, num_turns=2, num_tools=1, failed=0, completion_tokens=17.5]
step 6: 6%|▋ | 2/32 [00:01<00:38, 1.29s/it, reward=0.25, num_turns=2, num_tools=1, failed=0, completion_tokens=20]
step 6: 9%|▉ | 3/32 [00:01<00:37, 1.29s/it, reward=-0.5, num_turns=2, num_tools=1, failed=0, completion_tokens=20.2]
step 6: 12%|█▎ | 4/32 [00:01<00:36, 1.29s/it, reward=-0.125, num_turns=2, num_tools=1, failed=0, completion_tokens=21.1]
step 6: 16%|█▌ | 5/32 [00:01<00:34, 1.29s/it, reward=0.5, num_turns=2, num_tools=1, failed=0, completion_tokens=22.2]
step 6: 19%|█▉ | 6/32 [00:01<00:33, 1.29s/it, reward=0.833, num_turns=2, num_tools=1, failed=0, completion_tokens=22.9]
step 6: 22%|██▏ | 7/32 [00:01<00:32, 1.29s/it, reward=1.14, num_turns=2, num_tools=1, failed=0, completion_tokens=23.6]
step 6: 25%|██▌ | 8/32 [00:01<00:03, 7.66it/s, reward=1.14, num_turns=2, num_tools=1, failed=0, completion_tokens=23.6]
step 6: 25%|██▌ | 8/32 [00:01<00:03, 7.66it/s, reward=1.5, num_turns=2, num_tools=1, failed=0, completion_tokens=23.4]
step 6: 28%|██▊ | 9/32 [00:01<00:03, 7.66it/s, reward=1.78, num_turns=2, num_tools=1, failed=0, completion_tokens=23.2]
step 6: 31%|███▏ | 10/32 [00:01<00:02, 7.66it/s, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=23.1]
step 6: 34%|███▍ | 11/32 [00:01<00:02, 7.66it/s, reward=2.18, num_turns=2, num_tools=1, failed=0, completion_tokens=23.1]
step 6: 38%|███▊ | 12/32 [00:01<00:02, 7.66it/s, reward=2.33, num_turns=2, num_tools=1, failed=0, completion_tokens=23.1]
step 6: 41%|████ | 13/32 [00:01<00:02, 7.66it/s, reward=2.46, num_turns=2, num_tools=1, failed=0, completion_tokens=23.2]
step 6: 44%|████▍ | 14/32 [00:01<00:02, 7.66it/s, reward=2.46, num_turns=2, num_tools=1, failed=0, completion_tokens=23.8]
step 6: 47%|████▋ | 15/32 [00:01<00:02, 7.66it/s, reward=2.17, num_turns=2, num_tools=1, failed=0, completion_tokens=24.3]
step 6: 50%|█████ | 16/32 [00:01<00:02, 7.66it/s, reward=2.19, num_turns=2, num_tools=1, failed=0, completion_tokens=24.9]
step 6: 53%|█████▎ | 17/32 [00:01<00:01, 7.66it/s, reward=2.21, num_turns=2, num_tools=1, failed=0, completion_tokens=25.6]
step 6: 56%|█████▋ | 18/32 [00:01<00:01, 7.66it/s, reward=1.97, num_turns=2, num_tools=1, failed=0, completion_tokens=26.3]
step 6: 59%|█████▉ | 19/32 [00:01<00:00, 20.30it/s, reward=1.97, num_turns=2, num_tools=1, failed=0, completion_tokens=26.3]
step 6: 59%|█████▉ | 19/32 [00:01<00:00, 20.30it/s, reward=1.76, num_turns=2, num_tools=1, failed=0, completion_tokens=26.9]
step 6: 62%|██████▎ | 20/32 [00:01<00:00, 20.30it/s, reward=1.52, num_turns=1.95, num_tools=0.95, failed=0, completion_tokens=30.8]
step 6: 66%|██████▌ | 21/32 [00:01<00:00, 20.30it/s, reward=1.36, num_turns=1.95, num_tools=0.952, failed=0, completion_tokens=31.5]
step 6: 69%|██████▉ | 22/32 [00:01<00:00, 20.30it/s, reward=1.41, num_turns=1.95, num_tools=0.955, failed=0, completion_tokens=32.2]
step 6: 72%|███████▏ | 23/32 [00:01<00:00, 20.30it/s, reward=1.26, num_turns=1.96, num_tools=0.957, failed=0, completion_tokens=32.9]
step 6: 75%|███████▌ | 24/32 [00:01<00:00, 20.30it/s, reward=1.31, num_turns=1.96, num_tools=0.958, failed=0, completion_tokens=32.7]
step 6: 78%|███████▊ | 25/32 [00:01<00:00, 20.30it/s, reward=1.18, num_turns=1.96, num_tools=0.96, failed=0, completion_tokens=32.9]
step 6: 81%|████████▏ | 26/32 [00:01<00:00, 20.30it/s, reward=1.06, num_turns=1.96, num_tools=0.962, failed=0, completion_tokens=34.2]
step 6: 84%|████████▍ | 27/32 [00:01<00:00, 20.12it/s, reward=1.06, num_turns=1.96, num_tools=0.962, failed=0, completion_tokens=34.2]
step 6: 84%|████████▍ | 27/32 [00:01<00:00, 20.12it/s, reward=1.08, num_turns=1.96, num_tools=0.963, failed=0, completion_tokens=36]
step 6: 88%|████████▊ | 28/32 [00:02<00:00, 20.12it/s, reward=0.97, num_turns=1.96, num_tools=0.964, failed=0, completion_tokens=38.6]
step 6: 91%|█████████ | 29/32 [00:02<00:00, 20.12it/s, reward=0.868, num_turns=1.97, num_tools=0.966, failed=0, completion_tokens=43.2]
step 6: 94%|█████████▍| 30/32 [00:02<00:00, 20.12it/s, reward=0.739, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=54.4]
step 6: 97%|█████████▋| 31/32 [00:02<00:00, 20.12it/s, reward=0.618, num_turns=1.9, num_tools=0.903, failed=0, completion_tokens=64]
step 6: 100%|██████████| 32/32 [00:03<00:00, 20.12it/s, reward=0.505, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=78]
step 6: 100%|██████████| 32/32 [00:03<00:00, 9.77it/s, reward=0.505, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=78]
group 0: mean=+1.94 std=1.488 min=-2.0 max=+2.5 | Which country has a larger population, France or I
group 1: mean=-1.79 std=1.374 min=-3.0 max=+1.7 | What is the population of Germany divided by its a
group 2: mean=-1.88 std=1.166 min=-3.0 max=+1.0 | What is the population of Japan divided by its are
group 3: mean=+3.75 std=0.433 min=+3.0 max=+4.0 | What's the weather like in Dubai?
Avg reward: 0.505 | Avg tools/rollout: 0.9 | groups with variance: 4/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0004
Packed 32 trajectories into 4 sequences of length 2048
train: 0%| | 0/4 [00:00<?, ?it/s]
train: 25%|██▌ | 1/4 [00:02<00:06, 2.13s/it]
train: 25%|██▌ | 1/4 [00:02<00:06, 2.13s/it, loss/train=0.0339, loss/grad_norm=4.55, loss/learning_rate=5e-5, loss/entropy=0.78]
train: 50%|█████ | 2/4 [00:02<00:02, 1.08s/it, loss/train=0.0339, loss/grad_norm=4.55, loss/learning_rate=5e-5, loss/entropy=0.78]
train: 50%|█████ | 2/4 [00:02<00:02, 1.08s/it, loss/train=-0.308, loss/grad_norm=2.06, loss/learning_rate=5e-5, loss/entropy=1.54]
train: 75%|███████▌ | 3/4 [00:02<00:00, 1.34it/s, loss/train=-0.308, loss/grad_norm=2.06, loss/learning_rate=5e-5, loss/entropy=1.54]
train: 75%|███████▌ | 3/4 [00:02<00:00, 1.34it/s, loss/train=0.463, loss/grad_norm=2.5, loss/learning_rate=5e-5, loss/entropy=1.18]
train: 100%|██████████| 4/4 [00:03<00:00, 1.71it/s, loss/train=0.463, loss/grad_norm=2.5, loss/learning_rate=5e-5, loss/entropy=1.18]
train: 100%|██████████| 4/4 [00:03<00:00, 1.71it/s, loss/train=-0.826, loss/grad_norm=0.638, loss/learning_rate=5e-5, loss/entropy=0.878](APIServer pid=12946) Adapters before cleanup: ['default']
(APIServer pid=12946) Keeping active adapter(s): ['default']
(APIServer pid=12946) Adapters after cleanup: ['default']
train: 100%|██████████| 4/4 [00:31<00:00, 7.79s/it, loss/train=-0.826, loss/grad_norm=0.638, loss/learning_rate=5e-5, loss/entropy=0.878]
============================================================
Step 7/50
============================================================
step 7: 0%| | 0/32 [00:00<?, ?it/s]
step 7: 3%|▎ | 1/32 [00:01<00:37, 1.20s/it]
step 7: 3%|▎ | 1/32 [00:01<00:37, 1.20s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=34]
step 7: 6%|▋ | 2/32 [00:01<00:35, 1.20s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=38.5]
step 7: 9%|▉ | 3/32 [00:01<00:10, 2.70it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=38.5]
step 7: 9%|▉ | 3/32 [00:01<00:10, 2.70it/s, reward=-1.17, num_turns=1.33, num_tools=0.333, failed=0, completion_tokens=30.3]
step 7: 12%|█▎ | 4/32 [00:01<00:10, 2.70it/s, reward=-1.62, num_turns=1.25, num_tools=0.25, failed=0, completion_tokens=38]
step 7: 16%|█▌ | 5/32 [00:01<00:09, 2.70it/s, reward=-0.8, num_turns=1.4, num_tools=0.4, failed=0, completion_tokens=34]
step 7: 19%|█▉ | 6/32 [00:01<00:09, 2.70it/s, reward=-1, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=31.9]
step 7: 22%|██▏ | 7/32 [00:01<00:09, 2.70it/s, reward=-0.5, num_turns=1.57, num_tools=0.571, failed=0, completion_tokens=30.6]
step 7: 25%|██▌ | 8/32 [00:01<00:08, 2.70it/s, reward=-0.125, num_turns=1.62, num_tools=0.625, failed=0, completion_tokens=29.9]
step 7: 28%|██▊ | 9/32 [00:01<00:02, 9.80it/s, reward=-0.125, num_turns=1.62, num_tools=0.625, failed=0, completion_tokens=29.9]
step 7: 28%|██▊ | 9/32 [00:01<00:02, 9.80it/s, reward=-0.333, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=29.5]
step 7: 31%|███▏ | 10/32 [00:01<00:02, 9.80it/s, reward=-0.5, num_turns=1.7, num_tools=0.7, failed=0, completion_tokens=29.3]
step 7: 34%|███▍ | 11/32 [00:01<00:02, 9.80it/s, reward=-0.227, num_turns=1.73, num_tools=0.727, failed=0, completion_tokens=29.1]
step 7: 38%|███▊ | 12/32 [00:01<00:02, 9.80it/s, reward=-0.375, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=29.2]
step 7: 41%|████ | 13/32 [00:01<00:01, 9.80it/s, reward=-0.154, num_turns=1.77, num_tools=0.769, failed=0, completion_tokens=29.3]
step 7: 44%|████▍ | 14/32 [00:01<00:01, 9.80it/s, reward=0.0357, num_turns=1.79, num_tools=0.786, failed=0, completion_tokens=29.5]
step 7: 47%|████▋ | 15/32 [00:01<00:01, 9.80it/s, reward=-0.1, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=30.1]
step 7: 50%|█████ | 16/32 [00:01<00:01, 9.80it/s, reward=-0.219, num_turns=1.81, num_tools=0.812, failed=0, completion_tokens=30.3]
step 7: 53%|█████▎ | 17/32 [00:01<00:01, 9.80it/s, reward=-0.324, num_turns=1.82, num_tools=0.824, failed=0, completion_tokens=30.6]
step 7: 56%|█████▋ | 18/32 [00:01<00:00, 20.22it/s, reward=-0.324, num_turns=1.82, num_tools=0.824, failed=0, completion_tokens=30.6]
step 7: 56%|█████▋ | 18/32 [00:01<00:00, 20.22it/s, reward=-0.167, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=31.1]
step 7: 59%|█████▉ | 19/32 [00:01<00:00, 20.22it/s, reward=-0.263, num_turns=1.84, num_tools=0.842, failed=0, completion_tokens=31.8]
step 7: 62%|██████▎ | 20/32 [00:01<00:00, 20.22it/s, reward=-0.35, num_turns=1.85, num_tools=0.85, failed=0, completion_tokens=32.4]
step 7: 66%|██████▌ | 21/32 [00:01<00:00, 20.22it/s, reward=-0.429, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=33.1]
step 7: 69%|██████▉ | 22/32 [00:01<00:00, 20.22it/s, reward=-0.341, num_turns=1.86, num_tools=0.864, failed=0, completion_tokens=33.8]
step 7: 72%|███████▏ | 23/32 [00:01<00:00, 20.22it/s, reward=-0.217, num_turns=1.87, num_tools=0.87, failed=0, completion_tokens=33.7]
step 7: 75%|███████▌ | 24/32 [00:01<00:00, 25.32it/s, reward=-0.217, num_turns=1.87, num_tools=0.87, failed=0, completion_tokens=33.7]
step 7: 75%|███████▌ | 24/32 [00:01<00:00, 25.32it/s, reward=-0.125, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=34.9]
step 7: 78%|███████▊ | 25/32 [00:01<00:00, 25.32it/s, reward=-0.02, num_turns=1.88, num_tools=0.88, failed=0, completion_tokens=36]
step 7: 81%|████████▏ | 26/32 [00:01<00:00, 25.32it/s, reward=0.0192, num_turns=1.88, num_tools=0.885, failed=0, completion_tokens=37.1]
step 7: 84%|████████▍ | 27/32 [00:01<00:00, 25.32it/s, reward=0.111, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=36.8]
step 7: 88%|████████▊ | 28/32 [00:01<00:00, 25.32it/s, reward=0.0357, num_turns=1.89, num_tools=0.893, failed=0, completion_tokens=36.7]
step 7: 91%|█████████ | 29/32 [00:02<00:00, 22.16it/s, reward=0.0357, num_turns=1.89, num_tools=0.893, failed=0, completion_tokens=36.7]
step 7: 91%|█████████ | 29/32 [00:02<00:00, 22.16it/s, reward=-0.0345, num_turns=1.9, num_tools=0.897, failed=0, completion_tokens=38.4]
step 7: 94%|█████████▍| 30/32 [00:02<00:00, 22.16it/s, reward=-0.133, num_turns=1.87, num_tools=0.867, failed=0, completion_tokens=48.6]
step 7: 97%|█████████▋| 31/32 [00:02<00:00, 22.16it/s, reward=-0.226, num_turns=1.84, num_tools=0.839, failed=0, completion_tokens=58.2]
step 7: 100%|██████████| 32/32 [00:02<00:00, 22.16it/s, reward=-0.141, num_turns=1.84, num_tools=0.844, failed=0, completion_tokens=61.6]
step 7: 100%|██████████| 32/32 [00:02<00:00, 11.90it/s, reward=-0.141, num_turns=1.84, num_tools=0.844, failed=0, completion_tokens=61.6]
group 0: mean=+1.12 std=2.382 min=-3.0 max=+2.5 | Which country has a larger population, Japan or Br
group 1: mean=-0.94 std=1.944 min=-3.0 max=+2.0 | What is the distance from Earth to the Sun in km i
group 2: mean=+1.38 std=1.949 min=-2.0 max=+2.5 | Which country has a larger population, France or B
group 3: mean=-2.12 std=0.331 min=-3.0 max=-2.0 | What is the GDP of Germany?
Avg reward: -0.141 | Avg tools/rollout: 0.8 | groups with variance: 4/4
Packed 32 trajectories into 3 sequences of length 2048
train: 0%| | 0/3 [00:00<?, ?it/s]
train: 33%|███▎ | 1/3 [00:02<00:04, 2.11s/it]
train: 33%|███▎ | 1/3 [00:02<00:04, 2.11s/it, loss/train=-0.371, loss/grad_norm=3.5, loss/learning_rate=5e-5, loss/entropy=1.56]
train: 67%|██████▋ | 2/3 [00:02<00:01, 1.07s/it, loss/train=-0.371, loss/grad_norm=3.5, loss/learning_rate=5e-5, loss/entropy=1.56]
train: 67%|██████▋ | 2/3 [00:02<00:01, 1.07s/it, loss/train=0.0586, loss/grad_norm=0.753, loss/learning_rate=5e-5, loss/entropy=0.649]
train: 100%|██████████| 3/3 [00:02<00:00, 1.34it/s, loss/train=0.0586, loss/grad_norm=0.753, loss/learning_rate=5e-5, loss/entropy=0.649]
train: 100%|██████████| 3/3 [00:02<00:00, 1.34it/s, loss/train=-0.702, loss/grad_norm=2.84, loss/learning_rate=5e-5, loss/entropy=1.2] (APIServer pid=12946) Adapters before cleanup: ['default']
(APIServer pid=12946) Keeping active adapter(s): ['default']
(APIServer pid=12946) Adapters after cleanup: ['default']
train: 100%|██████████| 3/3 [00:30<00:00, 10.26s/it, loss/train=-0.702, loss/grad_norm=2.84, loss/learning_rate=5e-5, loss/entropy=1.2]
============================================================
Step 8/50
============================================================
step 8: 0%| | 0/32 [00:00<?, ?it/s]
step 8: 3%|▎ | 1/32 [00:00<00:30, 1.00it/s]
step 8: 3%|▎ | 1/32 [00:00<00:30, 1.00it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=1]
step 8: 6%|▋ | 2/32 [00:01<00:16, 1.81it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=1]
step 8: 6%|▋ | 2/32 [00:01<00:16, 1.81it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=19.5]
step 8: 9%|▉ | 3/32 [00:01<00:16, 1.81it/s, reward=-1.33, num_turns=1.33, num_tools=0.333, failed=0, completion_tokens=19.3]
step 8: 12%|█▎ | 4/32 [00:01<00:15, 1.81it/s, reward=-0.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=19.4]
step 8: 16%|█▌ | 5/32 [00:01<00:04, 5.45it/s, reward=-0.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=19.4]
step 8: 16%|█▌ | 5/32 [00:01<00:04, 5.45it/s, reward=-0.1, num_turns=1.6, num_tools=0.6, failed=0, completion_tokens=19.7]
step 8: 19%|█▉ | 6/32 [00:01<00:04, 5.45it/s, reward=0.167, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=20.2]
step 8: 22%|██▏ | 7/32 [00:01<00:04, 5.45it/s, reward=0.429, num_turns=1.71, num_tools=0.714, failed=0, completion_tokens=21]
step 8: 25%|██▌ | 8/32 [00:01<00:04, 5.45it/s, reward=0.625, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=21.3]
step 8: 28%|██▊ | 9/32 [00:01<00:04, 5.45it/s, reward=0.778, num_turns=1.78, num_tools=0.778, failed=0, completion_tokens=21.6]
step 8: 31%|███▏ | 10/32 [00:01<00:04, 5.45it/s, reward=0.5, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=22.1]
step 8: 34%|███▍ | 11/32 [00:01<00:03, 5.45it/s, reward=0.5, num_turns=1.82, num_tools=0.818, failed=0, completion_tokens=22.5]
step 8: 38%|███▊ | 12/32 [00:01<00:03, 5.45it/s, reward=0.292, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=23]
step 8: 41%|████ | 13/32 [00:01<00:03, 5.45it/s, reward=0.115, num_turns=1.85, num_tools=0.846, failed=0, completion_tokens=23.4]
step 8: 44%|████▍ | 14/32 [00:01<00:03, 5.45it/s, reward=-0.107, num_turns=1.79, num_tools=0.786, failed=0, completion_tokens=27.2]
step 8: 47%|████▋ | 15/32 [00:01<00:03, 5.45it/s, reward=-0.233, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=27.3]
step 8: 50%|█████ | 16/32 [00:01<00:00, 21.15it/s, reward=-0.233, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=27.3]
step 8: 50%|█████ | 16/32 [00:01<00:00, 21.15it/s, reward=-0.344, num_turns=1.81, num_tools=0.812, failed=0, completion_tokens=27]
step 8: 53%|█████▎ | 17/32 [00:01<00:00, 21.15it/s, reward=-0.441, num_turns=1.82, num_tools=0.824, failed=0, completion_tokens=27.6]
step 8: 56%|█████▋ | 18/32 [00:01<00:00, 21.15it/s, reward=-0.528, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=28.2]
step 8: 59%|█████▉ | 19/32 [00:01<00:00, 21.15it/s, reward=-0.474, num_turns=1.84, num_tools=0.842, failed=0, completion_tokens=28]
step 8: 62%|██████▎ | 20/32 [00:01<00:00, 21.15it/s, reward=-0.425, num_turns=1.85, num_tools=0.85, failed=0, completion_tokens=27.9]
step 8: 66%|██████▌ | 21/32 [00:01<00:00, 21.15it/s, reward=-0.381, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=27.9]
step 8: 69%|██████▉ | 22/32 [00:01<00:00, 21.15it/s, reward=-0.341, num_turns=1.86, num_tools=0.864, failed=0, completion_tokens=27.9]
step 8: 72%|███████▏ | 23/32 [00:01<00:00, 21.15it/s, reward=-0.413, num_turns=1.87, num_tools=0.87, failed=0, completion_tokens=28.5]
step 8: 75%|███████▌ | 24/32 [00:01<00:00, 29.53it/s, reward=-0.413, num_turns=1.87, num_tools=0.87, failed=0, completion_tokens=28.5]
step 8: 75%|███████▌ | 24/32 [00:01<00:00, 29.53it/s, reward=-0.479, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=29.2]
step 8: 78%|███████▊ | 25/32 [00:01<00:00, 29.53it/s, reward=-0.54, num_turns=1.88, num_tools=0.88, failed=0, completion_tokens=30.1]
step 8: 81%|████████▏ | 26/32 [00:01<00:00, 29.53it/s, reward=-0.596, num_turns=1.88, num_tools=0.885, failed=0, completion_tokens=30.2]
step 8: 84%|████████▍ | 27/32 [00:01<00:00, 29.53it/s, reward=-0.648, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=31.4]
step 8: 88%|████████▊ | 28/32 [00:01<00:00, 29.53it/s, reward=-0.696, num_turns=1.89, num_tools=0.893, failed=0, completion_tokens=32.6]
step 8: 91%|█████████ | 29/32 [00:01<00:00, 23.73it/s, reward=-0.696, num_turns=1.89, num_tools=0.893, failed=0, completion_tokens=32.6]
step 8: 91%|█████████ | 29/32 [00:01<00:00, 23.73it/s, reward=-0.741, num_turns=1.9, num_tools=0.897, failed=0, completion_tokens=33.9]
step 8: 94%|█████████▍| 30/32 [00:02<00:00, 23.73it/s, reward=-0.783, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=33.9]
step 8: 97%|█████████▋| 31/32 [00:02<00:00, 23.73it/s, reward=-0.855, num_turns=1.87, num_tools=0.871, failed=0, completion_tokens=41.5]
step 8: 100%|██████████| 32/32 [00:02<00:00, 23.73it/s, reward=-0.797, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=45.8]
step 8: 100%|██████████| 32/32 [00:02<00:00, 11.77it/s, reward=-0.797, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=45.8]
group 0: mean=+0.94 std=0.583 min=+0.5 max=+2.0 | What is the temperature in Tokyo in Fahrenheit?
group 1: mean=-0.75 std=2.165 min=-3.0 max=+2.0 | Which is hotter right now, Paris or Mumbai?
group 2: mean=-1.50 std=1.323 min=-2.0 max=+2.0 | Which is hotter right now, Tokyo or Dubai?
group 3: mean=-1.88 std=1.166 min=-3.0 max=+1.0 | How old was Guido van Rossum in 2020?
Avg reward: -0.797 | Avg tools/rollout: 0.9 | groups with variance: 4/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0006
Packed 32 trajectories into 3 sequences of length 2048
train: 0%| | 0/3 [00:00<?, ?it/s]
train: 33%|███▎ | 1/3 [00:02<00:04, 2.06s/it]
train: 33%|███▎ | 1/3 [00:02<00:04, 2.06s/it, loss/train=-0.724, loss/grad_norm=3.31, loss/learning_rate=5e-5, loss/entropy=0.82]
train: 67%|██████▋ | 2/3 [00:02<00:01, 1.06s/it, loss/train=-0.724, loss/grad_norm=3.31, loss/learning_rate=5e-5, loss/entropy=0.82]
train: 67%|██████▋ | 2/3 [00:02<00:01, 1.06s/it, loss/train=-3.07, loss/grad_norm=38, loss/learning_rate=5e-5, loss/entropy=0.91]
train: 100%|██████████| 3/3 [00:02<00:00, 1.36it/s, loss/train=-3.07, loss/grad_norm=38, loss/learning_rate=5e-5, loss/entropy=0.91]
train: 100%|██████████| 3/3 [00:02<00:00, 1.36it/s, loss/train=1.03, loss/grad_norm=0.954, loss/learning_rate=5e-5, loss/entropy=1] (APIServer pid=12946) Adapters before cleanup: ['default']
(APIServer pid=12946) Keeping active adapter(s): ['default']
(APIServer pid=12946) Adapters after cleanup: ['default']
[2026-04-13 02:28:34] ERROR base_events.py:1821: Task exception was never retrieved
future: <Task finished name='Task-2' coro=<LocalBackend._monitor_openai_server() done, defined at /usr/local/lib/python3.12/dist-packages/art/local/backend.py:416> exception=NotFoundError("Error code: 404 - {'error': {'message': 'The model `qwen-0.5b-tool-agent@8` does not exist.', 'type': 'NotFoundError', 'param': 'model', 'code': 404}}")>
Traceback (most recent call last):
File "/usr/lib/python3.12/asyncio/tasks.py", line 314, in __step_run_and_handle_result
result = coro.send(None)
^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/art/local/backend.py", line 468, in _monitor_openai_server
raise e
File "/usr/local/lib/python3.12/dist-packages/art/local/backend.py", line 453, in _monitor_openai_server
await openai_client.completions.create(
File "/usr/local/lib/python3.12/dist-packages/openai/resources/completions.py", line 1109, in create
return await self._post(
^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/openai/_base_client.py", line 1884, in post
return await self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/openai/_base_client.py", line 1669, in request
raise self._make_status_error_from_response(err.response) from None
openai.NotFoundError: Error code: 404 - {'error': {'message': 'The model `qwen-0.5b-tool-agent@8` does not exist.', 'type': 'NotFoundError', 'param': 'model', 'code': 404}}
Traceback (most recent call last):
File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "/workspace/RL-Trained-Tool-Use-Agent/src/train.py", line 171, in <module>
main()
File "/workspace/RL-Trained-Tool-Use-Agent/src/train.py", line 167, in main
asyncio.run(train(**kwargs))
File "/usr/local/lib/python3.12/dist-packages/nest_asyncio.py", line 30, in run
return loop.run_until_complete(task)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/nest_asyncio.py", line 98, in run_until_complete
return f.result()
^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/futures.py", line 203, in result
raise self._exception.with_traceback(self._exception_tb)
File "/usr/lib/python3.12/asyncio/tasks.py", line 316, in __step_run_and_handle_result
result = coro.throw(exc)
^^^^^^^^^^^^^^^
File "/workspace/RL-Trained-Tool-Use-Agent/src/train.py", line 113, in train
result = await backend.train(model, train_groups, learning_rate=learning_rate)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/art/local/backend.py", line 644, in train
async for metrics in self._train_model(
File "/usr/local/lib/python3.12/dist-packages/art/local/backend.py", line 783, in _train_model
async for result in service.train(
File "/usr/local/lib/python3.12/dist-packages/mp_actors/move.py", line 226, in async_gen_wrapper
send_value = yield await asyncio.wrap_future(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/futures.py", line 287, in __await__
yield self # This tells Task to wait for completion.
^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/tasks.py", line 385, in __wakeup
future.result()
File "/usr/lib/python3.12/asyncio/futures.py", line 203, in result
raise self._exception.with_traceback(self._exception_tb)
RuntimeError: Proxy is closing
train: 100%|██████████| 3/3 [00:32<00:00, 10.79s/it, loss/train=1.03, loss/grad_norm=0.954, loss/learning_rate=5e-5, loss/entropy=1]
Skipping import of cpp extensions due to incompatible torch version 2.10.0+cu128 for torchao version 0.15.0 Please see https://github.com/pytorch/ao/issues/2919 for more info
Loaded 200 train, 50 val scenarios
GRPO config: 4 scenarios/step × 8 rollouts/scenario = 32 rollouts/step
Skipping import of cpp extensions due to incompatible torch version 2.10.0+cu128 for torchao version 0.15.0 Please see https://github.com/pytorch/ao/issues/2919 for more info
/usr/local/lib/python3.12/dist-packages/art/__init__.py:37: UserWarning: WARNING: Unsloth should be imported before [transformers] to ensure all optimizations are applied. Your code may run slower or encounter memory issues without these optimizations.
Please restructure your imports with 'import unsloth' at the top of your file.
import unsloth # noqa: F401
🦥 Unsloth: Will patch your computer to enable 2x faster free finetuning.
🦥 Unsloth Zoo will now patch everything to make training faster!
==((====))== Unsloth 2026.3.3: Fast Qwen2 patching. Transformers: 5.2.0. vLLM: 0.17.0+art1.
\\ /| NVIDIA A100-SXM4-80GB. Num GPUs = 1. Max memory: 79.252 GB. Platform: Linux.
O^O/ \_/ \ Torch: 2.10.0+cu128. CUDA: 8.0. CUDA Toolkit: 12.8. Triton: 3.6.0
\ / Bfloat16 = TRUE. FA [Xformers = 0.0.35. FA2 = False]
"-____-" Free license: http://github.com/unslothai/unsloth
Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!
Loading weights: 0%| | 0/290 [00:00<?, ?it/s]
Loading weights: 0%| | 1/290 [00:00<00:00, 5714.31it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 0%| | 1/290 [00:00<00:00, 3672.77it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/290 [00:00<00:54, 5.30it/s, Materializing param=model.embed_tokens.weight]
Loading weights: 1%| | 2/290 [00:00<00:54, 5.30it/s, Materializing param=model.layers.0.input_layernorm.weight]
Loading weights: 1%| | 2/290 [00:00<00:54, 5.30it/s, Materializing param=model.layers.0.input_layernorm.weight]
Loading weights: 1%| | 3/290 [00:00<00:54, 5.30it/s, Materializing param=model.layers.0.mlp.down_proj.weight]
Loading weights: 1%| | 3/290 [00:00<00:54, 5.30it/s, Materializing param=model.layers.0.mlp.down_proj.weight]
Loading weights: 1%|▏ | 4/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.mlp.gate_proj.weight]
Loading weights: 1%|▏ | 4/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.mlp.gate_proj.weight]
Loading weights: 2%|▏ | 5/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.mlp.up_proj.weight]
Loading weights: 2%|▏ | 5/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.mlp.up_proj.weight]
Loading weights: 2%|▏ | 6/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.post_attention_layernorm.weight]
Loading weights: 2%|▏ | 6/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.post_attention_layernorm.weight]
Loading weights: 2%|▏ | 7/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.self_attn.k_proj.bias]
Loading weights: 2%|▏ | 7/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.self_attn.k_proj.bias]
Loading weights: 3%|▎ | 8/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.self_attn.k_proj.weight]
Loading weights: 3%|▎ | 8/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.self_attn.k_proj.weight]
Loading weights: 3%|▎ | 9/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.self_attn.o_proj.weight]
Loading weights: 3%|▎ | 9/290 [00:00<00:53, 5.30it/s, Materializing param=model.layers.0.self_attn.o_proj.weight]
Loading weights: 3%|▎ | 10/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.0.self_attn.q_proj.bias]
Loading weights: 3%|▎ | 10/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.0.self_attn.q_proj.bias]
Loading weights: 4%|▍ | 11/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.0.self_attn.q_proj.weight]
Loading weights: 4%|▍ | 11/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.0.self_attn.q_proj.weight]
Loading weights: 4%|▍ | 12/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.0.self_attn.v_proj.bias]
Loading weights: 4%|▍ | 12/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.0.self_attn.v_proj.bias]
Loading weights: 4%|▍ | 13/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.0.self_attn.v_proj.weight]
Loading weights: 4%|▍ | 13/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.0.self_attn.v_proj.weight]
Loading weights: 5%|▍ | 14/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.1.input_layernorm.weight]
Loading weights: 5%|▍ | 14/290 [00:00<00:52, 5.30it/s, Materializing param=model.layers.1.input_layernorm.weight]
Loading weights: 5%|▌ | 15/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.mlp.down_proj.weight]
Loading weights: 5%|▌ | 15/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.mlp.down_proj.weight]
Loading weights: 6%|▌ | 16/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.mlp.gate_proj.weight]
Loading weights: 6%|▌ | 16/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.mlp.gate_proj.weight]
Loading weights: 6%|▌ | 17/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.mlp.up_proj.weight]
Loading weights: 6%|▌ | 17/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.mlp.up_proj.weight]
Loading weights: 6%|▌ | 18/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.post_attention_layernorm.weight]
Loading weights: 6%|▌ | 18/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.post_attention_layernorm.weight]
Loading weights: 7%|▋ | 19/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.self_attn.k_proj.bias]
Loading weights: 7%|▋ | 19/290 [00:00<00:51, 5.30it/s, Materializing param=model.layers.1.self_attn.k_proj.bias]
Loading weights: 7%|▋ | 20/290 [00:00<00:50, 5.30it/s, Materializing param=model.layers.1.self_attn.k_proj.weight]
Loading weights: 7%|▋ | 20/290 [00:00<00:50, 5.30it/s, Materializing param=model.layers.1.self_attn.k_proj.weight]
Loading weights: 7%|▋ | 21/290 [00:00<00:50, 5.30it/s, Materializing param=model.layers.1.self_attn.o_proj.weight]
Loading weights: 7%|▋ | 21/290 [00:00<00:50, 5.30it/s, Materializing param=model.layers.1.self_attn.o_proj.weight]
Loading weights: 8%|▊ | 22/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.o_proj.weight]
Loading weights: 8%|▊ | 22/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.q_proj.bias]
Loading weights: 8%|▊ | 22/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.q_proj.bias]
Loading weights: 8%|▊ | 23/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.q_proj.weight]
Loading weights: 8%|▊ | 23/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.q_proj.weight]
Loading weights: 8%|▊ | 24/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.v_proj.bias]
Loading weights: 8%|▊ | 24/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.v_proj.bias]
Loading weights: 9%|▊ | 25/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.v_proj.weight]
Loading weights: 9%|▊ | 25/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.1.self_attn.v_proj.weight]
Loading weights: 9%|▉ | 26/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.input_layernorm.weight]
Loading weights: 9%|▉ | 26/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.input_layernorm.weight]
Loading weights: 9%|▉ | 27/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.mlp.down_proj.weight]
Loading weights: 9%|▉ | 27/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.mlp.down_proj.weight]
Loading weights: 10%|▉ | 28/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.mlp.gate_proj.weight]
Loading weights: 10%|▉ | 28/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.mlp.gate_proj.weight]
Loading weights: 10%|█ | 29/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.mlp.up_proj.weight]
Loading weights: 10%|█ | 29/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.mlp.up_proj.weight]
Loading weights: 10%|█ | 30/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.post_attention_layernorm.weight]
Loading weights: 10%|█ | 30/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.post_attention_layernorm.weight]
Loading weights: 11%|█ | 31/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.k_proj.bias]
Loading weights: 11%|█ | 31/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.k_proj.bias]
Loading weights: 11%|█ | 32/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.k_proj.weight]
Loading weights: 11%|█ | 32/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.k_proj.weight]
Loading weights: 11%|█▏ | 33/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.o_proj.weight]
Loading weights: 11%|█▏ | 33/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.o_proj.weight]
Loading weights: 12%|█▏ | 34/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.q_proj.bias]
Loading weights: 12%|█▏ | 34/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.q_proj.bias]
Loading weights: 12%|█▏ | 35/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.q_proj.weight]
Loading weights: 12%|█▏ | 35/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.q_proj.weight]
Loading weights: 12%|█▏ | 36/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.v_proj.bias]
Loading weights: 12%|█▏ | 36/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.v_proj.bias]
Loading weights: 13%|█▎ | 37/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.v_proj.weight]
Loading weights: 13%|█▎ | 37/290 [00:00<00:04, 53.67it/s, Materializing param=model.layers.2.self_attn.v_proj.weight]
Loading weights: 13%|█▎ | 38/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.2.self_attn.v_proj.weight]
Loading weights: 13%|█▎ | 38/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.input_layernorm.weight]
Loading weights: 13%|█▎ | 38/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.input_layernorm.weight]
Loading weights: 13%|█▎ | 39/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.mlp.down_proj.weight]
Loading weights: 13%|█▎ | 39/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.mlp.down_proj.weight]
Loading weights: 14%|█▍ | 40/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.mlp.gate_proj.weight]
Loading weights: 14%|█▍ | 40/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.mlp.gate_proj.weight]
Loading weights: 14%|█▍ | 41/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.mlp.up_proj.weight]
Loading weights: 14%|█▍ | 41/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.mlp.up_proj.weight]
Loading weights: 14%|█▍ | 42/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.post_attention_layernorm.weight]
Loading weights: 14%|█▍ | 42/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.post_attention_layernorm.weight]
Loading weights: 15%|█▍ | 43/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.k_proj.bias]
Loading weights: 15%|█▍ | 43/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.k_proj.bias]
Loading weights: 15%|█▌ | 44/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.k_proj.weight]
Loading weights: 15%|█▌ | 44/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.k_proj.weight]
Loading weights: 16%|█▌ | 45/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.o_proj.weight]
Loading weights: 16%|█▌ | 45/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.o_proj.weight]
Loading weights: 16%|█▌ | 46/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.q_proj.bias]
Loading weights: 16%|█▌ | 46/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.q_proj.bias]
Loading weights: 16%|█▌ | 47/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.q_proj.weight]
Loading weights: 16%|█▌ | 47/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.q_proj.weight]
Loading weights: 17%|█▋ | 48/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.v_proj.bias]
Loading weights: 17%|█▋ | 48/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.v_proj.bias]
Loading weights: 17%|█▋ | 49/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.v_proj.weight]
Loading weights: 17%|█▋ | 49/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.3.self_attn.v_proj.weight]
Loading weights: 17%|█▋ | 50/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.4.input_layernorm.weight]
Loading weights: 17%|█▋ | 50/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.4.input_layernorm.weight]
Loading weights: 18%|█▊ | 51/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.4.mlp.down_proj.weight]
Loading weights: 18%|█▊ | 51/290 [00:00<00:03, 79.49it/s, Materializing param=model.layers.4.mlp.down_proj.weight]
Loading weights: 18%|█▊ | 52/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.mlp.gate_proj.weight]
Loading weights: 18%|█▊ | 52/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.mlp.gate_proj.weight]
Loading weights: 18%|█▊ | 53/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.mlp.up_proj.weight]
Loading weights: 18%|█▊ | 53/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.mlp.up_proj.weight]
Loading weights: 19%|█▊ | 54/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.post_attention_layernorm.weight]
Loading weights: 19%|█▊ | 54/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.post_attention_layernorm.weight]
Loading weights: 19%|█▉ | 55/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.k_proj.bias]
Loading weights: 19%|█▉ | 55/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.k_proj.bias]
Loading weights: 19%|█▉ | 56/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.k_proj.weight]
Loading weights: 19%|█▉ | 56/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.k_proj.weight]
Loading weights: 20%|█▉ | 57/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.o_proj.weight]
Loading weights: 20%|█▉ | 57/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.o_proj.weight]
Loading weights: 20%|██ | 58/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.q_proj.bias]
Loading weights: 20%|██ | 58/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.q_proj.bias]
Loading weights: 20%|██ | 59/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.q_proj.weight]
Loading weights: 20%|██ | 59/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.q_proj.weight]
Loading weights: 21%|██ | 60/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.v_proj.bias]
Loading weights: 21%|██ | 60/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.v_proj.bias]
Loading weights: 21%|██ | 61/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.v_proj.weight]
Loading weights: 21%|██ | 61/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.4.self_attn.v_proj.weight]
Loading weights: 21%|██▏ | 62/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.input_layernorm.weight]
Loading weights: 21%|██▏ | 62/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.input_layernorm.weight]
Loading weights: 22%|██▏ | 63/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.mlp.down_proj.weight]
Loading weights: 22%|██▏ | 63/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.mlp.down_proj.weight]
Loading weights: 22%|██▏ | 64/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.mlp.gate_proj.weight]
Loading weights: 22%|██▏ | 64/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.mlp.gate_proj.weight]
Loading weights: 22%|██▏ | 65/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.mlp.up_proj.weight]
Loading weights: 22%|██▏ | 65/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.mlp.up_proj.weight]
Loading weights: 23%|██▎ | 66/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.post_attention_layernorm.weight]
Loading weights: 23%|██▎ | 66/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.post_attention_layernorm.weight]
Loading weights: 23%|██▎ | 67/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.k_proj.bias]
Loading weights: 23%|██▎ | 67/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.k_proj.bias]
Loading weights: 23%|██▎ | 68/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.k_proj.weight]
Loading weights: 23%|██▎ | 68/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.k_proj.weight]
Loading weights: 24%|██▍ | 69/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.o_proj.weight]
Loading weights: 24%|██▍ | 69/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.o_proj.weight]
Loading weights: 24%|██▍ | 70/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.q_proj.bias]
Loading weights: 24%|██▍ | 70/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.q_proj.bias]
Loading weights: 24%|██▍ | 71/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.q_proj.weight]
Loading weights: 24%|██▍ | 71/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.q_proj.weight]
Loading weights: 25%|██▍ | 72/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.v_proj.bias]
Loading weights: 25%|██▍ | 72/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.v_proj.bias]
Loading weights: 25%|██▌ | 73/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.v_proj.weight]
Loading weights: 25%|██▌ | 73/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.5.self_attn.v_proj.weight]
Loading weights: 26%|██▌ | 74/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.6.input_layernorm.weight]
Loading weights: 26%|██▌ | 74/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.6.input_layernorm.weight]
Loading weights: 26%|██▌ | 75/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.6.mlp.down_proj.weight]
Loading weights: 26%|██▌ | 75/290 [00:00<00:02, 79.49it/s, Materializing param=model.layers.6.mlp.down_proj.weight]
Loading weights: 26%|██▌ | 76/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.mlp.down_proj.weight]
Loading weights: 26%|██▌ | 76/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.mlp.gate_proj.weight]
Loading weights: 26%|██▌ | 76/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.mlp.gate_proj.weight]
Loading weights: 27%|██▋ | 77/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.mlp.up_proj.weight]
Loading weights: 27%|██▋ | 77/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.mlp.up_proj.weight]
Loading weights: 27%|██▋ | 78/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.post_attention_layernorm.weight]
Loading weights: 27%|██▋ | 78/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.post_attention_layernorm.weight]
Loading weights: 27%|██▋ | 79/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.k_proj.bias]
Loading weights: 27%|██▋ | 79/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.k_proj.bias]
Loading weights: 28%|██▊ | 80/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.k_proj.weight]
Loading weights: 28%|██▊ | 80/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.k_proj.weight]
Loading weights: 28%|██▊ | 81/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.o_proj.weight]
Loading weights: 28%|██▊ | 81/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.o_proj.weight]
Loading weights: 28%|██▊ | 82/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.q_proj.bias]
Loading weights: 28%|██▊ | 82/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.q_proj.bias]
Loading weights: 29%|██▊ | 83/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.q_proj.weight]
Loading weights: 29%|██▊ | 83/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.q_proj.weight]
Loading weights: 29%|██▉ | 84/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.v_proj.bias]
Loading weights: 29%|██▉ | 84/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.v_proj.bias]
Loading weights: 29%|██▉ | 85/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.v_proj.weight]
Loading weights: 29%|██▉ | 85/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.6.self_attn.v_proj.weight]
Loading weights: 30%|██▉ | 86/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.input_layernorm.weight]
Loading weights: 30%|██▉ | 86/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.input_layernorm.weight]
Loading weights: 30%|███ | 87/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.mlp.down_proj.weight]
Loading weights: 30%|███ | 87/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.mlp.down_proj.weight]
Loading weights: 30%|███ | 88/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.mlp.gate_proj.weight]
Loading weights: 30%|███ | 88/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.mlp.gate_proj.weight]
Loading weights: 31%|███ | 89/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.mlp.up_proj.weight]
Loading weights: 31%|███ | 89/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.mlp.up_proj.weight]
Loading weights: 31%|███ | 90/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.post_attention_layernorm.weight]
Loading weights: 31%|███ | 90/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.post_attention_layernorm.weight]
Loading weights: 31%|███▏ | 91/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.k_proj.bias]
Loading weights: 31%|███▏ | 91/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.k_proj.bias]
Loading weights: 32%|███▏ | 92/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.k_proj.weight]
Loading weights: 32%|███▏ | 92/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.k_proj.weight]
Loading weights: 32%|███▏ | 93/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.o_proj.weight]
Loading weights: 32%|███▏ | 93/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.o_proj.weight]
Loading weights: 32%|███▏ | 94/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.q_proj.bias]
Loading weights: 32%|███▏ | 94/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.q_proj.bias]
Loading weights: 33%|███▎ | 95/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.q_proj.weight]
Loading weights: 33%|███▎ | 95/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.q_proj.weight]
Loading weights: 33%|███▎ | 96/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.v_proj.bias]
Loading weights: 33%|███▎ | 96/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.v_proj.bias]
Loading weights: 33%|███▎ | 97/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.v_proj.weight]
Loading weights: 33%|███▎ | 97/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.7.self_attn.v_proj.weight]
Loading weights: 34%|███▍ | 98/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.input_layernorm.weight]
Loading weights: 34%|███▍ | 98/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.input_layernorm.weight]
Loading weights: 34%|███▍ | 99/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.mlp.down_proj.weight]
Loading weights: 34%|███▍ | 99/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.mlp.down_proj.weight]
Loading weights: 34%|███▍ | 100/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.mlp.gate_proj.weight]
Loading weights: 34%|███▍ | 100/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.mlp.gate_proj.weight]
Loading weights: 35%|███▍ | 101/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.mlp.up_proj.weight]
Loading weights: 35%|███▍ | 101/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.mlp.up_proj.weight]
Loading weights: 35%|███▌ | 102/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.post_attention_layernorm.weight]
Loading weights: 35%|███▌ | 102/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.post_attention_layernorm.weight]
Loading weights: 36%|███▌ | 103/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.self_attn.k_proj.bias]
Loading weights: 36%|███▌ | 103/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.self_attn.k_proj.bias]
Loading weights: 36%|███▌ | 104/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.self_attn.k_proj.weight]
Loading weights: 36%|███▌ | 104/290 [00:00<00:01, 159.98it/s, Materializing param=model.layers.8.self_attn.k_proj.weight]
Loading weights: 36%|███▌ | 105/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.k_proj.weight]
Loading weights: 36%|███▌ | 105/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.o_proj.weight]
Loading weights: 36%|███▌ | 105/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.o_proj.weight]
Loading weights: 37%|███▋ | 106/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.q_proj.bias]
Loading weights: 37%|███▋ | 106/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.q_proj.bias]
Loading weights: 37%|███▋ | 107/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.q_proj.weight]
Loading weights: 37%|███▋ | 107/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.q_proj.weight]
Loading weights: 37%|███▋ | 108/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.v_proj.bias]
Loading weights: 37%|███▋ | 108/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.v_proj.bias]
Loading weights: 38%|███▊ | 109/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.v_proj.weight]
Loading weights: 38%|███▊ | 109/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.8.self_attn.v_proj.weight]
Loading weights: 38%|███▊ | 110/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.input_layernorm.weight]
Loading weights: 38%|███▊ | 110/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.input_layernorm.weight]
Loading weights: 38%|███▊ | 111/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.mlp.down_proj.weight]
Loading weights: 38%|███▊ | 111/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.mlp.down_proj.weight]
Loading weights: 39%|███▊ | 112/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.mlp.gate_proj.weight]
Loading weights: 39%|███▊ | 112/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.mlp.gate_proj.weight]
Loading weights: 39%|███▉ | 113/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.mlp.up_proj.weight]
Loading weights: 39%|███▉ | 113/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.mlp.up_proj.weight]
Loading weights: 39%|███▉ | 114/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.post_attention_layernorm.weight]
Loading weights: 39%|███▉ | 114/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.post_attention_layernorm.weight]
Loading weights: 40%|███▉ | 115/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.k_proj.bias]
Loading weights: 40%|███▉ | 115/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.k_proj.bias]
Loading weights: 40%|████ | 116/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.k_proj.weight]
Loading weights: 40%|████ | 116/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.k_proj.weight]
Loading weights: 40%|████ | 117/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.o_proj.weight]
Loading weights: 40%|████ | 117/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.o_proj.weight]
Loading weights: 41%|████ | 118/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.q_proj.bias]
Loading weights: 41%|████ | 118/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.q_proj.bias]
Loading weights: 41%|████ | 119/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 41%|████ | 119/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.q_proj.weight]
Loading weights: 41%|████▏ | 120/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.v_proj.bias]
Loading weights: 41%|████▏ | 120/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.v_proj.bias]
Loading weights: 42%|████▏ | 121/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.v_proj.weight]
Loading weights: 42%|████▏ | 121/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.9.self_attn.v_proj.weight]
Loading weights: 42%|████▏ | 122/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.input_layernorm.weight]
Loading weights: 42%|████▏ | 122/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.input_layernorm.weight]
Loading weights: 42%|████▏ | 123/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.mlp.down_proj.weight]
Loading weights: 42%|████▏ | 123/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.mlp.down_proj.weight]
Loading weights: 43%|████▎ | 124/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.mlp.gate_proj.weight]
Loading weights: 43%|████▎ | 124/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.mlp.gate_proj.weight]
Loading weights: 43%|████▎ | 125/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 43%|████▎ | 125/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.mlp.up_proj.weight]
Loading weights: 43%|████▎ | 126/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 43%|████▎ | 126/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.post_attention_layernorm.weight]
Loading weights: 44%|████▍ | 127/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.k_proj.bias]
Loading weights: 44%|████▍ | 127/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.k_proj.bias]
Loading weights: 44%|████▍ | 128/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.k_proj.weight]
Loading weights: 44%|████▍ | 128/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.k_proj.weight]
Loading weights: 44%|████▍ | 129/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.o_proj.weight]
Loading weights: 44%|████▍ | 129/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.o_proj.weight]
Loading weights: 45%|████▍ | 130/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.q_proj.bias]
Loading weights: 45%|████▍ | 130/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.q_proj.bias]
Loading weights: 45%|████▌ | 131/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.q_proj.weight]
Loading weights: 45%|████▌ | 131/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.q_proj.weight]
Loading weights: 46%|████▌ | 132/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.v_proj.bias]
Loading weights: 46%|████▌ | 132/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.v_proj.bias]
Loading weights: 46%|████▌ | 133/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.v_proj.weight]
Loading weights: 46%|████▌ | 133/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.10.self_attn.v_proj.weight]
Loading weights: 46%|████▌ | 134/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.input_layernorm.weight]
Loading weights: 46%|████▌ | 134/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.input_layernorm.weight]
Loading weights: 47%|████▋ | 135/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.mlp.down_proj.weight]
Loading weights: 47%|████▋ | 135/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.mlp.down_proj.weight]
Loading weights: 47%|████▋ | 136/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.mlp.gate_proj.weight]
Loading weights: 47%|████▋ | 136/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.mlp.gate_proj.weight]
Loading weights: 47%|████▋ | 137/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.mlp.up_proj.weight]
Loading weights: 47%|████▋ | 137/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.mlp.up_proj.weight]
Loading weights: 48%|████▊ | 138/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.post_attention_layernorm.weight]
Loading weights: 48%|████▊ | 138/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.post_attention_layernorm.weight]
Loading weights: 48%|████▊ | 139/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.self_attn.k_proj.bias]
Loading weights: 48%|████▊ | 139/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.self_attn.k_proj.bias]
Loading weights: 48%|████▊ | 140/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.self_attn.k_proj.weight]
Loading weights: 48%|████▊ | 140/290 [00:00<00:00, 194.21it/s, Materializing param=model.layers.11.self_attn.k_proj.weight]
Loading weights: 49%|████▊ | 141/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.k_proj.weight]
Loading weights: 49%|████▊ | 141/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.o_proj.weight]
Loading weights: 49%|████▊ | 141/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.o_proj.weight]
Loading weights: 49%|████▉ | 142/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.q_proj.bias]
Loading weights: 49%|████▉ | 142/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.q_proj.bias]
Loading weights: 49%|████▉ | 143/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.q_proj.weight]
Loading weights: 49%|████▉ | 143/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.q_proj.weight]
Loading weights: 50%|████▉ | 144/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.v_proj.bias]
Loading weights: 50%|████▉ | 144/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.v_proj.bias]
Loading weights: 50%|█████ | 145/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 145/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.11.self_attn.v_proj.weight]
Loading weights: 50%|█████ | 146/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.input_layernorm.weight]
Loading weights: 50%|█████ | 146/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.input_layernorm.weight]
Loading weights: 51%|█████ | 147/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.mlp.down_proj.weight]
Loading weights: 51%|█████ | 147/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.mlp.down_proj.weight]
Loading weights: 51%|█████ | 148/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.mlp.gate_proj.weight]
Loading weights: 51%|█████ | 148/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.mlp.gate_proj.weight]
Loading weights: 51%|█████▏ | 149/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.mlp.up_proj.weight]
Loading weights: 51%|█████▏ | 149/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.mlp.up_proj.weight]
Loading weights: 52%|█████▏ | 150/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 150/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.post_attention_layernorm.weight]
Loading weights: 52%|█████▏ | 151/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.k_proj.bias]
Loading weights: 52%|█████▏ | 151/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.k_proj.bias]
Loading weights: 52%|█████▏ | 152/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.k_proj.weight]
Loading weights: 52%|█████▏ | 152/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.k_proj.weight]
Loading weights: 53%|█████▎ | 153/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.o_proj.weight]
Loading weights: 53%|█████▎ | 153/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.o_proj.weight]
Loading weights: 53%|█████▎ | 154/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.q_proj.bias]
Loading weights: 53%|█████▎ | 154/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.q_proj.bias]
Loading weights: 53%|█████▎ | 155/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.q_proj.weight]
Loading weights: 53%|█████▎ | 155/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.q_proj.weight]
Loading weights: 54%|█████▍ | 156/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.v_proj.bias]
Loading weights: 54%|█████▍ | 156/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.v_proj.bias]
Loading weights: 54%|█████▍ | 157/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.v_proj.weight]
Loading weights: 54%|█████▍ | 157/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.12.self_attn.v_proj.weight]
Loading weights: 54%|█████▍ | 158/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.input_layernorm.weight]
Loading weights: 54%|█████▍ | 158/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.input_layernorm.weight]
Loading weights: 55%|█████▍ | 159/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.mlp.down_proj.weight]
Loading weights: 55%|█████▍ | 159/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.mlp.down_proj.weight]
Loading weights: 55%|█████▌ | 160/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.mlp.gate_proj.weight]
Loading weights: 55%|█████▌ | 160/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.mlp.gate_proj.weight]
Loading weights: 56%|█████▌ | 161/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.mlp.up_proj.weight]
Loading weights: 56%|█████▌ | 161/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.mlp.up_proj.weight]
Loading weights: 56%|█████▌ | 162/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 162/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.post_attention_layernorm.weight]
Loading weights: 56%|█████▌ | 163/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.k_proj.bias]
Loading weights: 56%|█████▌ | 163/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.k_proj.bias]
Loading weights: 57%|█████▋ | 164/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.k_proj.weight]
Loading weights: 57%|█████▋ | 164/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.k_proj.weight]
Loading weights: 57%|█████▋ | 165/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.o_proj.weight]
Loading weights: 57%|█████▋ | 165/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.o_proj.weight]
Loading weights: 57%|█████▋ | 166/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.q_proj.bias]
Loading weights: 57%|█████▋ | 166/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.q_proj.bias]
Loading weights: 58%|█████▊ | 167/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 58%|█████▊ | 167/290 [00:00<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.q_proj.weight]
Loading weights: 58%|█████▊ | 168/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.v_proj.bias]
Loading weights: 58%|█████▊ | 168/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.v_proj.bias]
Loading weights: 58%|█████▊ | 169/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 58%|█████▊ | 169/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.13.self_attn.v_proj.weight]
Loading weights: 59%|█████▊ | 170/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 59%|█████▊ | 170/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.input_layernorm.weight]
Loading weights: 59%|█████▉ | 171/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 59%|█████▉ | 171/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.mlp.down_proj.weight]
Loading weights: 59%|█████▉ | 172/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.mlp.gate_proj.weight]
Loading weights: 59%|█████▉ | 172/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.mlp.gate_proj.weight]
Loading weights: 60%|█████▉ | 173/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 60%|█████▉ | 173/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.mlp.up_proj.weight]
Loading weights: 60%|██████ | 174/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 60%|██████ | 174/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.post_attention_layernorm.weight]
Loading weights: 60%|██████ | 175/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.self_attn.k_proj.bias]
Loading weights: 60%|██████ | 175/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.self_attn.k_proj.bias]
Loading weights: 61%|██████ | 176/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.self_attn.k_proj.weight]
Loading weights: 61%|██████ | 176/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.self_attn.k_proj.weight]
Loading weights: 61%|██████ | 177/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.self_attn.o_proj.weight]
Loading weights: 61%|██████ | 177/290 [00:01<00:00, 240.51it/s, Materializing param=model.layers.14.self_attn.o_proj.weight]
Loading weights: 61%|██████▏ | 178/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.o_proj.weight]
Loading weights: 61%|██████▏ | 178/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.q_proj.bias]
Loading weights: 61%|██████▏ | 178/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.q_proj.bias]
Loading weights: 62%|██████▏ | 179/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.q_proj.weight]
Loading weights: 62%|██████▏ | 179/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.q_proj.weight]
Loading weights: 62%|██████▏ | 180/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.v_proj.bias]
Loading weights: 62%|██████▏ | 180/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.v_proj.bias]
Loading weights: 62%|██████▏ | 181/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.v_proj.weight]
Loading weights: 62%|██████▏ | 181/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.14.self_attn.v_proj.weight]
Loading weights: 63%|██████▎ | 182/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.input_layernorm.weight]
Loading weights: 63%|██████▎ | 182/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.input_layernorm.weight]
Loading weights: 63%|██████▎ | 183/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.mlp.down_proj.weight]
Loading weights: 63%|██████▎ | 183/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.mlp.down_proj.weight]
Loading weights: 63%|██████▎ | 184/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.mlp.gate_proj.weight]
Loading weights: 63%|██████▎ | 184/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.mlp.gate_proj.weight]
Loading weights: 64%|██████▍ | 185/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.mlp.up_proj.weight]
Loading weights: 64%|██████▍ | 185/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.mlp.up_proj.weight]
Loading weights: 64%|██████▍ | 186/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 64%|██████▍ | 186/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.post_attention_layernorm.weight]
Loading weights: 64%|██████▍ | 187/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.k_proj.bias]
Loading weights: 64%|██████▍ | 187/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.k_proj.bias]
Loading weights: 65%|██████▍ | 188/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 65%|██████▍ | 188/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.k_proj.weight]
Loading weights: 65%|██████▌ | 189/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 65%|██████▌ | 189/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.o_proj.weight]
Loading weights: 66%|██████▌ | 190/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.q_proj.bias]
Loading weights: 66%|██████▌ | 190/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.q_proj.bias]
Loading weights: 66%|██████▌ | 191/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.q_proj.weight]
Loading weights: 66%|██████▌ | 191/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.q_proj.weight]
Loading weights: 66%|██████▌ | 192/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.v_proj.bias]
Loading weights: 66%|██████▌ | 192/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.v_proj.bias]
Loading weights: 67%|██████▋ | 193/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.v_proj.weight]
Loading weights: 67%|██████▋ | 193/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.15.self_attn.v_proj.weight]
Loading weights: 67%|██████▋ | 194/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.input_layernorm.weight]
Loading weights: 67%|██████▋ | 194/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.input_layernorm.weight]
Loading weights: 67%|██████▋ | 195/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.mlp.down_proj.weight]
Loading weights: 67%|██████▋ | 195/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.mlp.down_proj.weight]
Loading weights: 68%|██████▊ | 196/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.mlp.gate_proj.weight]
Loading weights: 68%|██████▊ | 196/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.mlp.gate_proj.weight]
Loading weights: 68%|██████▊ | 197/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.mlp.up_proj.weight]
Loading weights: 68%|██████▊ | 197/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.mlp.up_proj.weight]
Loading weights: 68%|██████▊ | 198/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.post_attention_layernorm.weight]
Loading weights: 68%|██████▊ | 198/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.post_attention_layernorm.weight]
Loading weights: 69%|██████▊ | 199/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.k_proj.bias]
Loading weights: 69%|██████▊ | 199/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.k_proj.bias]
Loading weights: 69%|██████▉ | 200/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.k_proj.weight]
Loading weights: 69%|██████▉ | 200/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.k_proj.weight]
Loading weights: 69%|██████▉ | 201/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.o_proj.weight]
Loading weights: 69%|██████▉ | 201/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.o_proj.weight]
Loading weights: 70%|██████▉ | 202/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.q_proj.bias]
Loading weights: 70%|██████▉ | 202/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.q_proj.bias]
Loading weights: 70%|███████ | 203/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.q_proj.weight]
Loading weights: 70%|███████ | 203/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.q_proj.weight]
Loading weights: 70%|███████ | 204/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.v_proj.bias]
Loading weights: 70%|███████ | 204/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.v_proj.bias]
Loading weights: 71%|███████ | 205/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.v_proj.weight]
Loading weights: 71%|███████ | 205/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.16.self_attn.v_proj.weight]
Loading weights: 71%|███████ | 206/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.17.input_layernorm.weight]
Loading weights: 71%|███████ | 206/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.17.input_layernorm.weight]
Loading weights: 71%|███████▏ | 207/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.17.mlp.down_proj.weight]
Loading weights: 71%|███████▏ | 207/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.17.mlp.down_proj.weight]
Loading weights: 72%|███████▏ | 208/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.17.mlp.gate_proj.weight]
Loading weights: 72%|███████▏ | 208/290 [00:01<00:00, 275.02it/s, Materializing param=model.layers.17.mlp.gate_proj.weight]
Loading weights: 72%|███████▏ | 209/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.mlp.gate_proj.weight]
Loading weights: 72%|███████▏ | 209/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 72%|███████▏ | 209/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.mlp.up_proj.weight]
Loading weights: 72%|███████▏ | 210/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 72%|███████▏ | 210/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.post_attention_layernorm.weight]
Loading weights: 73%|███████▎ | 211/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.k_proj.bias]
Loading weights: 73%|███████▎ | 211/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.k_proj.bias]
Loading weights: 73%|███████▎ | 212/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.k_proj.weight]
Loading weights: 73%|███████▎ | 212/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.k_proj.weight]
Loading weights: 73%|███████▎ | 213/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.o_proj.weight]
Loading weights: 73%|███████▎ | 213/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.o_proj.weight]
Loading weights: 74%|███████▍ | 214/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.q_proj.bias]
Loading weights: 74%|███████▍ | 214/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.q_proj.bias]
Loading weights: 74%|███████▍ | 215/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.q_proj.weight]
Loading weights: 74%|███████▍ | 215/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.q_proj.weight]
Loading weights: 74%|███████▍ | 216/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.v_proj.bias]
Loading weights: 74%|███████▍ | 216/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.v_proj.bias]
Loading weights: 75%|███████▍ | 217/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.v_proj.weight]
Loading weights: 75%|███████▍ | 217/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.17.self_attn.v_proj.weight]
Loading weights: 75%|███████▌ | 218/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.input_layernorm.weight]
Loading weights: 75%|███████▌ | 218/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.input_layernorm.weight]
Loading weights: 76%|███████▌ | 219/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.mlp.down_proj.weight]
Loading weights: 76%|███████▌ | 219/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.mlp.down_proj.weight]
Loading weights: 76%|███████▌ | 220/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.mlp.gate_proj.weight]
Loading weights: 76%|███████▌ | 220/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.mlp.gate_proj.weight]
Loading weights: 76%|███████▌ | 221/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.mlp.up_proj.weight]
Loading weights: 76%|███████▌ | 221/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.mlp.up_proj.weight]
Loading weights: 77%|███████▋ | 222/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 222/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.post_attention_layernorm.weight]
Loading weights: 77%|███████▋ | 223/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.k_proj.bias]
Loading weights: 77%|███████▋ | 223/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.k_proj.bias]
Loading weights: 77%|███████▋ | 224/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.k_proj.weight]
Loading weights: 77%|███████▋ | 224/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.k_proj.weight]
Loading weights: 78%|███████▊ | 225/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.o_proj.weight]
Loading weights: 78%|███████▊ | 225/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.o_proj.weight]
Loading weights: 78%|███████▊ | 226/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.q_proj.bias]
Loading weights: 78%|███████▊ | 226/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.q_proj.bias]
Loading weights: 78%|███████▊ | 227/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.q_proj.weight]
Loading weights: 78%|███████▊ | 227/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.q_proj.weight]
Loading weights: 79%|███████▊ | 228/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.v_proj.bias]
Loading weights: 79%|███████▊ | 228/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.v_proj.bias]
Loading weights: 79%|███████▉ | 229/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.v_proj.weight]
Loading weights: 79%|███████▉ | 229/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.18.self_attn.v_proj.weight]
Loading weights: 79%|███████▉ | 230/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.input_layernorm.weight]
Loading weights: 79%|███████▉ | 230/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.input_layernorm.weight]
Loading weights: 80%|███████▉ | 231/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.mlp.down_proj.weight]
Loading weights: 80%|███████▉ | 231/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.mlp.down_proj.weight]
Loading weights: 80%|████████ | 232/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.mlp.gate_proj.weight]
Loading weights: 80%|████████ | 232/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.mlp.gate_proj.weight]
Loading weights: 80%|████████ | 233/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.mlp.up_proj.weight]
Loading weights: 80%|████████ | 233/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.mlp.up_proj.weight]
Loading weights: 81%|████████ | 234/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.post_attention_layernorm.weight]
Loading weights: 81%|████████ | 234/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.post_attention_layernorm.weight]
Loading weights: 81%|████████ | 235/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.k_proj.bias]
Loading weights: 81%|████████ | 235/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.k_proj.bias]
Loading weights: 81%|████████▏ | 236/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.k_proj.weight]
Loading weights: 81%|████████▏ | 236/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.k_proj.weight]
Loading weights: 82%|████████▏ | 237/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.o_proj.weight]
Loading weights: 82%|████████▏ | 237/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.o_proj.weight]
Loading weights: 82%|████████▏ | 238/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.q_proj.bias]
Loading weights: 82%|████████▏ | 238/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.q_proj.bias]
Loading weights: 82%|████████▏ | 239/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.q_proj.weight]
Loading weights: 82%|████████▏ | 239/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.q_proj.weight]
Loading weights: 83%|████████▎ | 240/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.v_proj.bias]
Loading weights: 83%|████████▎ | 240/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.v_proj.bias]
Loading weights: 83%|████████▎ | 241/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.v_proj.weight]
Loading weights: 83%|████████▎ | 241/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.19.self_attn.v_proj.weight]
Loading weights: 83%|████████▎ | 242/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.20.input_layernorm.weight]
Loading weights: 83%|████████▎ | 242/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.20.input_layernorm.weight]
Loading weights: 84%|████████▍ | 243/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.20.mlp.down_proj.weight]
Loading weights: 84%|████████▍ | 243/290 [00:01<00:00, 280.26it/s, Materializing param=model.layers.20.mlp.down_proj.weight]
Loading weights: 84%|████████▍ | 244/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.mlp.down_proj.weight]
Loading weights: 84%|████████▍ | 244/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.mlp.gate_proj.weight]
Loading weights: 84%|████████▍ | 244/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.mlp.gate_proj.weight]
Loading weights: 84%|████████▍ | 245/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.mlp.up_proj.weight]
Loading weights: 84%|████████▍ | 245/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.mlp.up_proj.weight]
Loading weights: 85%|████████▍ | 246/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.post_attention_layernorm.weight]
Loading weights: 85%|████████▍ | 246/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.post_attention_layernorm.weight]
Loading weights: 85%|████████▌ | 247/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.k_proj.bias]
Loading weights: 85%|████████▌ | 247/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.k_proj.bias]
Loading weights: 86%|████████▌ | 248/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.k_proj.weight]
Loading weights: 86%|████████▌ | 248/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.k_proj.weight]
Loading weights: 86%|████████▌ | 249/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.o_proj.weight]
Loading weights: 86%|████████▌ | 249/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.o_proj.weight]
Loading weights: 86%|████████▌ | 250/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.q_proj.bias]
Loading weights: 86%|████████▌ | 250/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.q_proj.bias]
Loading weights: 87%|████████▋ | 251/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.q_proj.weight]
Loading weights: 87%|████████▋ | 251/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.q_proj.weight]
Loading weights: 87%|████████▋ | 252/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.v_proj.bias]
Loading weights: 87%|████████▋ | 252/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.v_proj.bias]
Loading weights: 87%|████████▋ | 253/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.v_proj.weight]
Loading weights: 87%|████████▋ | 253/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.20.self_attn.v_proj.weight]
Loading weights: 88%|████████▊ | 254/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.input_layernorm.weight]
Loading weights: 88%|████████▊ | 254/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.input_layernorm.weight]
Loading weights: 88%|████████▊ | 255/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.mlp.down_proj.weight]
Loading weights: 88%|████████▊ | 255/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.mlp.down_proj.weight]
Loading weights: 88%|████████▊ | 256/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.mlp.gate_proj.weight]
Loading weights: 88%|████████▊ | 256/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.mlp.gate_proj.weight]
Loading weights: 89%|████████▊ | 257/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 89%|████████▊ | 257/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.mlp.up_proj.weight]
Loading weights: 89%|████████▉ | 258/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 89%|████████▉ | 258/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.post_attention_layernorm.weight]
Loading weights: 89%|████████▉ | 259/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.k_proj.bias]
Loading weights: 89%|████████▉ | 259/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.k_proj.bias]
Loading weights: 90%|████████▉ | 260/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.k_proj.weight]
Loading weights: 90%|████████▉ | 260/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.k_proj.weight]
Loading weights: 90%|█████████ | 261/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.o_proj.weight]
Loading weights: 90%|█████████ | 261/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.o_proj.weight]
Loading weights: 90%|█████████ | 262/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.q_proj.bias]
Loading weights: 90%|█████████ | 262/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.q_proj.bias]
Loading weights: 91%|█████████ | 263/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.q_proj.weight]
Loading weights: 91%|█████████ | 263/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.q_proj.weight]
Loading weights: 91%|█████████ | 264/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.v_proj.bias]
Loading weights: 91%|█████████ | 264/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.v_proj.bias]
Loading weights: 91%|█████████▏| 265/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.v_proj.weight]
Loading weights: 91%|█████████▏| 265/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.21.self_attn.v_proj.weight]
Loading weights: 92%|█████████▏| 266/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.input_layernorm.weight]
Loading weights: 92%|█████████▏| 266/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.input_layernorm.weight]
Loading weights: 92%|█████████▏| 267/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.mlp.down_proj.weight]
Loading weights: 92%|█████████▏| 267/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.mlp.down_proj.weight]
Loading weights: 92%|█████████▏| 268/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.mlp.gate_proj.weight]
Loading weights: 92%|█████████▏| 268/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.mlp.gate_proj.weight]
Loading weights: 93%|█████████▎| 269/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.mlp.up_proj.weight]
Loading weights: 93%|█████████▎| 269/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.mlp.up_proj.weight]
Loading weights: 93%|█████████▎| 270/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 93%|█████████▎| 270/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.post_attention_layernorm.weight]
Loading weights: 93%|█████████▎| 271/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.k_proj.bias]
Loading weights: 93%|█████████▎| 271/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.k_proj.bias]
Loading weights: 94%|█████████▍| 272/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.k_proj.weight]
Loading weights: 94%|█████████▍| 272/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.k_proj.weight]
Loading weights: 94%|█████████▍| 273/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.o_proj.weight]
Loading weights: 94%|█████████▍| 273/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.o_proj.weight]
Loading weights: 94%|█████████▍| 274/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.q_proj.bias]
Loading weights: 94%|█████████▍| 274/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.q_proj.bias]
Loading weights: 95%|█████████▍| 275/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.q_proj.weight]
Loading weights: 95%|█████████▍| 275/290 [00:01<00:00, 298.67it/s, Materializing param=model.layers.22.self_attn.q_proj.weight]
Loading weights: 95%|█████████▌| 276/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.22.self_attn.q_proj.weight]
Loading weights: 95%|█████████▌| 276/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.22.self_attn.v_proj.bias]
Loading weights: 95%|█████████▌| 276/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.22.self_attn.v_proj.bias]
Loading weights: 96%|█████████▌| 277/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.22.self_attn.v_proj.weight]
Loading weights: 96%|█████████▌| 277/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.22.self_attn.v_proj.weight]
Loading weights: 96%|█████████▌| 278/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.input_layernorm.weight]
Loading weights: 96%|█████████▌| 278/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.input_layernorm.weight]
Loading weights: 96%|█████████▌| 279/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.mlp.down_proj.weight]
Loading weights: 96%|█████████▌| 279/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.mlp.down_proj.weight]
Loading weights: 97%|█████████▋| 280/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 97%|█████████▋| 280/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.mlp.gate_proj.weight]
Loading weights: 97%|█████████▋| 281/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 97%|█████████▋| 281/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.mlp.up_proj.weight]
Loading weights: 97%|█████████▋| 282/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 97%|█████████▋| 282/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.post_attention_layernorm.weight]
Loading weights: 98%|█████████▊| 283/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.k_proj.bias]
Loading weights: 98%|█████████▊| 283/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.k_proj.bias]
Loading weights: 98%|█████████▊| 284/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.k_proj.weight]
Loading weights: 98%|█████████▊| 284/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.k_proj.weight]
Loading weights: 98%|█████████▊| 285/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.o_proj.weight]
Loading weights: 98%|█████████▊| 285/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.o_proj.weight]
Loading weights: 99%|█████████▊| 286/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.q_proj.bias]
Loading weights: 99%|█████████▊| 286/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.q_proj.bias]
Loading weights: 99%|█████████▉| 287/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.q_proj.weight]
Loading weights: 99%|█████████▉| 287/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.q_proj.weight]
Loading weights: 99%|█████████▉| 288/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.v_proj.bias]
Loading weights: 99%|█████████▉| 288/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.v_proj.bias]
Loading weights: 100%|█████████▉| 289/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.v_proj.weight]
Loading weights: 100%|█████████▉| 289/290 [00:01<00:00, 276.47it/s, Materializing param=model.layers.23.self_attn.v_proj.weight]
Loading weights: 100%|██████████| 290/290 [00:01<00:00, 276.47it/s, Materializing param=model.norm.weight]
Loading weights: 100%|██████████| 290/290 [00:01<00:00, 276.47it/s, Materializing param=model.norm.weight]
Loading weights: 100%|██████████| 290/290 [00:01<00:00, 206.03it/s, Materializing param=model.norm.weight]
unsloth/qwen2.5-0.5b-instruct-unsloth-bnb-4bit does not have a padding token! Will use pad_token = <|PAD_TOKEN|>.
Unsloth 2026.3.3 patched 24 layers with 24 QKV layers, 24 O layers and 24 MLP layers.
Warning: `huggingface-cli` is deprecated and no longer works. Use `hf` instead.

Hint: `hf` is already installed! Use it directly.

Hint: Examples:
hf auth login
hf download unsloth/gemma-4-31B-it-GGUF
hf upload my-cool-model . .
hf models ls --search "gemma"
hf repos ls --format json
hf jobs run python:3.12 python -c 'print("Hello!")'
hf --help

INFO 04-13 02:36:48 [model.py:531] Resolved architecture: Qwen2ForCausalLM
INFO 04-13 02:36:48 [model.py:1554] Using max model len 32768
INFO 04-13 02:36:48 [scheduler.py:231] Chunked prefill is enabled with max_num_batched_tokens=2048.
INFO 04-13 02:36:48 [vllm.py:747] Asynchronous scheduling is enabled.
WARNING 04-13 02:36:50 [system_utils.py:152] We must use the `spawn` multiprocessing start method. Overriding VLLM_WORKER_MULTIPROC_METHOD to 'spawn'. See https://docs.vllm.ai/en/latest/usage/troubleshooting.html#python-multiprocessing for more information. Reasons: CUDA is initialized
Skipping import of cpp extensions due to incompatible torch version 2.10.0+cu128 for torchao version 0.15.0 Please see https://github.com/pytorch/ao/issues/2919 for more info
/usr/local/lib/python3.12/dist-packages/art/__init__.py:37: UserWarning: WARNING: Unsloth should be imported before [transformers] to ensure all optimizations are applied. Your code may run slower or encounter memory issues without these optimizations.
Please restructure your imports with 'import unsloth' at the top of your file.
import unsloth # noqa: F401
🦥 Unsloth: Will patch your computer to enable 2x faster free finetuning.
🦥 Unsloth Zoo will now patch everything to make training faster!
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:15 [core.py:101] Initializing a V1 LLM engine (v0.17.0+art1) with config: model='Qwen/Qwen2.5-0.5B-Instruct', speculative_config=None, tokenizer='Qwen/Qwen2.5-0.5B-Instruct', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, tokenizer_revision=None, trust_remote_code=False, dtype=torch.bfloat16, max_seq_len=32768, download_dir=None, load_format=auto, tensor_parallel_size=1, pipeline_parallel_size=1, data_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, enable_return_routed_experts=False, kv_cache_dtype=auto, device_config=cuda, structured_outputs_config=StructuredOutputsConfig(backend='auto', disable_fallback=False, disable_any_whitespace=False, disable_additional_properties=False, reasoning_parser='', reasoning_parser_plugin='', enable_in_reasoning=False), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None, kv_cache_metrics=False, kv_cache_metrics_sample=0.01, cudagraph_metrics=False, enable_layerwise_nvtx_tracing=False, enable_mfu_metrics=False, enable_mm_processor_stats=False, enable_logging_iteration_details=False), seed=0, served_model_name=Qwen/Qwen2.5-0.5B-Instruct, enable_prefix_caching=True, enable_chunked_prefill=True, pooler_config=None, compilation_config={'level': None, 'mode': <CompilationMode.VLLM_COMPILE: 3>, 'debug_dump_path': None, 'cache_dir': '', 'compile_cache_save_format': 'binary', 'backend': 'inductor', 'custom_ops': ['none'], 'splitting_ops': ['vllm::unified_attention', 'vllm::unified_attention_with_output', 'vllm::unified_mla_attention', 'vllm::unified_mla_attention_with_output', 'vllm::mamba_mixer2', 'vllm::mamba_mixer', 'vllm::short_conv', 'vllm::linear_attention', 'vllm::plamo2_mamba_mixer', 'vllm::gdn_attention_core', 'vllm::kda_attention', 'vllm::sparse_attn_indexer', 'vllm::rocm_aiter_sparse_attn_indexer', 'vllm::unified_kv_cache_update', 'vllm::unified_mla_kv_cache_update'], 'compile_mm_encoder': False, 'compile_sizes': [], 'compile_ranges_split_points': [2048], 'inductor_compile_config': {'enable_auto_functionalized_v2': False, 'combo_kernels': True, 'benchmark_combo_kernel': True}, 'inductor_passes': {}, 'cudagraph_mode': <CUDAGraphMode.FULL_AND_PIECEWISE: (2, 1)>, 'cudagraph_num_of_warmups': 1, 'cudagraph_capture_sizes': [1, 2, 4, 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256], 'cudagraph_copy_inputs': False, 'cudagraph_specialize_lora': True, 'use_inductor_graph_partition': False, 'pass_config': {'fuse_norm_quant': False, 'fuse_act_quant': False, 'fuse_attn_quant': False, 'enable_sp': False, 'fuse_gemm_comms': False, 'fuse_allreduce_rms': False}, 'max_cudagraph_capture_size': 256, 'dynamic_shapes_config': {'type': <DynamicShapesType.BACKED: 'backed'>, 'evaluate_guards': False, 'assume_32_bit_indexing': False}, 'local_cache_dir': None, 'fast_moe_cold_start': True, 'static_all_moe_layers': []}
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:15 [worker_base.py:283] Injected <class 'art.vllm.engine.WorkerExtension'> into <class 'vllm.v1.worker.gpu_worker.Worker'> for extended collective_rpc calls ['run', 'time']
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:15 [parallel_state.py:1393] world_size=1 rank=0 local_rank=0 distributed_init_method=tcp://172.21.0.2:42797 backend=nccl
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:15 [parallel_state.py:1715] rank 0 in world size 1 is assigned as DP rank 0, PP rank 0, PCP rank 0, TP rank 0, EP rank N/A, EPLB rank N/A
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:16 [base.py:106] Offloader set to NoopOffloader
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:16 [gpu_model_runner.py:4255] Starting to load model Qwen/Qwen2.5-0.5B-Instruct...
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:17 [cuda.py:405] Using FLASH_ATTN attention backend out of potential backends: ['FLASH_ATTN', 'FLASHINFER', 'TRITON_ATTN', 'FLEX_ATTENTION'].
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:17 [flash_attn.py:587] Using FlashAttention version 2
(EngineCore_DP0 pid=15589) <frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
(EngineCore_DP0 pid=15589) <frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:18 [weight_utils.py:601] No model.safetensors.index.json found in remote.
(EngineCore_DP0 pid=15589)
Loading safetensors checkpoint shards: 0% Completed | 0/1 [00:00<?, ?it/s]
(EngineCore_DP0 pid=15589)
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:01<00:00, 1.78s/it]
(EngineCore_DP0 pid=15589)
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:01<00:00, 1.78s/it]
(EngineCore_DP0 pid=15589)
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:20 [default_loader.py:293] Loading weights took 1.78 seconds
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:20 [punica_selector.py:20] Using PunicaWrapperGPU.
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:21 [gpu_model_runner.py:4338] Model loading took 0.96 GiB memory and 3.620917 seconds
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:37 [decorators.py:465] Directly load AOT compilation from path /root/.cache/vllm/torch_compile_cache/torch_aot_compile/19f16ef5be162d523fe85c0ed27f944cf1ccd27d08e2ae363d4b7c12b35022cc/rank_0_0/model
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:37 [backends.py:916] Using cache directory: /root/.cache/vllm/torch_compile_cache/d97828e2e7/rank_0_0/backbone for vLLM's torch.compile
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:37 [backends.py:976] Dynamo bytecode transform time: 3.13 s
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:40 [backends.py:266] Directly load the compiled graph(s) for compile range (1, 2048) from the cache, took 1.439 s
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:40 [monitor.py:35] torch.compile takes 5.47 s in total
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:41 [gpu_worker.py:424] Available KV cache memory: 70.01 GiB
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:41 [kv_cache_utils.py:1314] GPU KV cache size: 6,117,600 tokens
(EngineCore_DP0 pid=15589) INFO 04-13 02:37:41 [kv_cache_utils.py:1319] Maximum concurrency for 32,768 tokens per request: 186.69x
(EngineCore_DP0 pid=15589)
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 0%| | 0/70 [00:00<?, ?it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 1%|▏ | 1/70 [00:12<14:40, 12.76s/it](EngineCore_DP0 pid=15589) WARNING 04-13 02:37:56 [utils.py:268] Using default LoRA kernel configs
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 3%|▎ | 2/70 [00:12<06:05, 5.37s/it]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 6%|▌ | 4/70 [00:13<02:14, 2.04s/it]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 9%|▊ | 6/70 [00:13<01:10, 1.11s/it]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 11%|█▏ | 8/70 [00:13<00:42, 1.46it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 14%|█▍ | 10/70 [00:13<00:27, 2.19it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 17%|█▋ | 12/70 [00:13<00:18, 3.10it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 20%|██ | 14/70 [00:13<00:13, 4.23it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 23%|██▎ | 16/70 [00:13<00:09, 5.55it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 26%|██▌ | 18/70 [00:13<00:07, 7.03it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 29%|██▊ | 20/70 [00:14<00:05, 8.58it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 31%|███▏ | 22/70 [00:14<00:04, 10.09it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 34%|███▍ | 24/70 [00:14<00:03, 11.52it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 37%|███▋ | 26/70 [00:14<00:03, 12.52it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 40%|████ | 28/70 [00:14<00:03, 13.39it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 43%|████▎ | 30/70 [00:14<00:02, 14.13it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 46%|████▌ | 32/70 [00:14<00:02, 14.39it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 49%|████▊ | 34/70 [00:14<00:02, 14.80it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 51%|█████▏ | 36/70 [00:15<00:02, 14.96it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 54%|█████▍ | 38/70 [00:15<00:02, 14.85it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 57%|█████▋ | 40/70 [00:15<00:01, 15.56it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 60%|██████ | 42/70 [00:15<00:01, 15.78it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 63%|██████▎ | 44/70 [00:15<00:01, 15.34it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 66%|██████▌ | 46/70 [00:15<00:01, 15.87it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 69%|██████▊ | 48/70 [00:15<00:01, 16.38it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 71%|███████▏ | 50/70 [00:15<00:01, 16.61it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 74%|███████▍ | 52/70 [00:16<00:01, 16.84it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 77%|███████▋ | 54/70 [00:16<00:00, 16.85it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 80%|████████ | 56/70 [00:16<00:00, 16.74it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 83%|████████▎ | 58/70 [00:16<00:00, 15.93it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 86%|████████▌ | 60/70 [00:16<00:00, 16.18it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 89%|████████▊ | 62/70 [00:16<00:00, 16.71it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 91%|█████████▏| 64/70 [00:16<00:00, 17.01it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 94%|█████████▍| 66/70 [00:16<00:00, 17.19it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 97%|█████████▋| 68/70 [00:16<00:00, 17.44it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 100%|██████████| 70/70 [00:17<00:00, 15.84it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 100%|██████████| 70/70 [00:17<00:00, 4.09it/s]
(EngineCore_DP0 pid=15589)
Capturing CUDA graphs (decode, FULL): 0%| | 0/38 [00:00<?, ?it/s]
Capturing CUDA graphs (decode, FULL): 3%|▎ | 1/38 [00:12<07:49, 12.69s/it]
Capturing CUDA graphs (decode, FULL): 11%|█ | 4/38 [00:12<01:23, 2.44s/it]
Capturing CUDA graphs (decode, FULL): 16%|█▌ | 6/38 [00:12<00:44, 1.40s/it]
Capturing CUDA graphs (decode, FULL): 21%|██ | 8/38 [00:13<00:26, 1.13it/s]
Capturing CUDA graphs (decode, FULL): 26%|██▋ | 10/38 [00:13<00:16, 1.69it/s]
Capturing CUDA graphs (decode, FULL): 32%|███▏ | 12/38 [00:13<00:10, 2.43it/s]
Capturing CUDA graphs (decode, FULL): 37%|███▋ | 14/38 [00:13<00:07, 3.38it/s]
Capturing CUDA graphs (decode, FULL): 42%|████▏ | 16/38 [00:13<00:04, 4.55it/s]
Capturing CUDA graphs (decode, FULL): 47%|████▋ | 18/38 [00:13<00:03, 5.91it/s]
Capturing CUDA graphs (decode, FULL): 53%|█████▎ | 20/38 [00:13<00:02, 7.39it/s]
Capturing CUDA graphs (decode, FULL): 58%|█████▊ | 22/38 [00:13<00:01, 8.96it/s]
Capturing CUDA graphs (decode, FULL): 63%|██████▎ | 24/38 [00:13<00:01, 10.47it/s]
Capturing CUDA graphs (decode, FULL): 68%|██████▊ | 26/38 [00:14<00:01, 11.88it/s]
Capturing CUDA graphs (decode, FULL): 74%|███████▎ | 28/38 [00:14<00:00, 13.17it/s]
Capturing CUDA graphs (decode, FULL): 79%|███████▉ | 30/38 [00:14<00:00, 14.22it/s]
Capturing CUDA graphs (decode, FULL): 84%|████████▍ | 32/38 [00:14<00:00, 15.04it/s]
Capturing CUDA graphs (decode, FULL): 89%|████████▉ | 34/38 [00:14<00:00, 15.68it/s]
Capturing CUDA graphs (decode, FULL): 95%|█████████▍| 36/38 [00:14<00:00, 16.10it/s]
Capturing CUDA graphs (decode, FULL): 100%|██████████| 38/38 [00:14<00:00, 16.44it/s]
Capturing CUDA graphs (decode, FULL): 100%|██████████| 38/38 [00:14<00:00, 2.57it/s]
(EngineCore_DP0 pid=15589) INFO 04-13 02:38:16 [gpu_model_runner.py:5360] Graph capturing finished in 33 secs, took 0.65 GiB
(EngineCore_DP0 pid=15589) INFO 04-13 02:38:28 [core.py:282] init engine (profile, create kv cache, warmup model) took 67.40 seconds
(EngineCore_DP0 pid=15589) INFO 04-13 02:38:32 [vllm.py:747] Asynchronous scheduling is enabled.
Starting from step 8
============================================================
Step 9/50
============================================================
step 9: 0%| | 0/32 [00:00<?, ?it/s]
step 9: 3%|▎ | 1/32 [00:01<01:01, 1.98s/it]
step 9: 3%|▎ | 1/32 [00:01<01:01, 1.98s/it, reward=3, num_turns=2, num_tools=1, failed=0, completion_tokens=20.5]
step 9: 6%|▋ | 2/32 [00:01<00:59, 1.98s/it, reward=2.25, num_turns=2, num_tools=1, failed=0, completion_tokens=21]
step 9: 9%|▉ | 3/32 [00:02<00:57, 1.98s/it, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=22]
step 9: 12%|█▎ | 4/32 [00:02<00:55, 1.98s/it, reward=1.62, num_turns=2, num_tools=1, failed=0, completion_tokens=22.9]
step 9: 16%|█▌ | 5/32 [00:02<00:08, 3.08it/s, reward=1.62, num_turns=2, num_tools=1, failed=0, completion_tokens=22.9]
step 9: 16%|█▌ | 5/32 [00:02<00:08, 3.08it/s, reward=1.4, num_turns=2, num_tools=1, failed=0, completion_tokens=23.7]
step 9: 19%|█▉ | 6/32 [00:02<00:08, 3.08it/s, reward=1.25, num_turns=2, num_tools=1, failed=0, completion_tokens=24.2]
step 9: 22%|██▏ | 7/32 [00:02<00:08, 3.08it/s, reward=1.64, num_turns=2, num_tools=1, failed=0, completion_tokens=24.9]
step 9: 25%|██▌ | 8/32 [00:02<00:07, 3.08it/s, reward=1.69, num_turns=2, num_tools=1, failed=0, completion_tokens=24.9]
step 9: 28%|██▊ | 9/32 [00:02<00:07, 3.08it/s, reward=1.94, num_turns=2, num_tools=1, failed=0, completion_tokens=25.3]
step 9: 31%|███▏ | 10/32 [00:02<00:07, 3.08it/s, reward=1.8, num_turns=2, num_tools=1, failed=0, completion_tokens=25.8]
step 9: 34%|███▍ | 11/32 [00:02<00:06, 3.08it/s, reward=1.68, num_turns=2, num_tools=1, failed=0, completion_tokens=26.1]
step 9: 38%|███▊ | 12/32 [00:02<00:06, 3.08it/s, reward=1.88, num_turns=2, num_tools=1, failed=0, completion_tokens=26.2]
step 9: 41%|████ | 13/32 [00:02<00:06, 3.08it/s, reward=2.04, num_turns=2, num_tools=1, failed=0, completion_tokens=26.3]
step 9: 44%|████▍ | 14/32 [00:02<00:05, 3.08it/s, reward=1.75, num_turns=2, num_tools=1, failed=0, completion_tokens=26.5]
step 9: 47%|████▋ | 15/32 [00:02<00:05, 3.08it/s, reward=1.9, num_turns=2, num_tools=1, failed=0, completion_tokens=26.6]
step 9: 50%|█████ | 16/32 [00:02<00:05, 3.08it/s, reward=1.66, num_turns=2, num_tools=1, failed=0, completion_tokens=26.7]
step 9: 53%|█████▎ | 17/32 [00:02<00:04, 3.08it/s, reward=1.75, num_turns=2, num_tools=1, failed=0, completion_tokens=26.8]
step 9: 56%|█████▋ | 18/32 [00:02<00:04, 3.08it/s, reward=1.55, num_turns=2, num_tools=1, failed=0, completion_tokens=27]
step 9: 59%|█████▉ | 19/32 [00:02<00:04, 3.08it/s, reward=1.31, num_turns=1.95, num_tools=0.947, failed=0, completion_tokens=30.1]
step 9: 62%|██████▎ | 20/32 [00:02<00:03, 3.08it/s, reward=1.14, num_turns=1.95, num_tools=0.95, failed=0, completion_tokens=30.1]
step 9: 66%|██████▌ | 21/32 [00:02<00:03, 3.08it/s, reward=0.992, num_turns=1.95, num_tools=0.952, failed=0, completion_tokens=30.2]
step 9: 69%|██████▉ | 22/32 [00:02<00:03, 3.08it/s, reward=0.856, num_turns=1.95, num_tools=0.955, failed=0, completion_tokens=30.4]
step 9: 72%|███████▏ | 23/32 [00:02<00:02, 3.08it/s, reward=0.688, num_turns=1.91, num_tools=0.913, failed=0, completion_tokens=33]
step 9: 75%|███████▌ | 24/32 [00:02<00:02, 3.08it/s, reward=0.826, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=32.8]
step 9: 78%|███████▊ | 25/32 [00:02<00:00, 19.61it/s, reward=0.826, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=32.8]
step 9: 78%|███████▊ | 25/32 [00:02<00:00, 19.61it/s, reward=0.713, num_turns=1.92, num_tools=0.92, failed=0, completion_tokens=32.9]
step 9: 81%|████████▏ | 26/32 [00:02<00:00, 19.61it/s, reward=0.609, num_turns=1.92, num_tools=0.923, failed=0, completion_tokens=33.3]
step 9: 84%|████████▍ | 27/32 [00:02<00:00, 19.61it/s, reward=0.512, num_turns=1.93, num_tools=0.926, failed=0, completion_tokens=33.3]
step 9: 88%|████████▊ | 28/32 [00:02<00:00, 19.61it/s, reward=0.637, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=33.1]
step 9: 91%|█████████ | 29/32 [00:02<00:00, 19.61it/s, reward=0.546, num_turns=1.93, num_tools=0.931, failed=0, completion_tokens=33.9]
step 9: 94%|█████████▍| 30/32 [00:02<00:00, 19.61it/s, reward=0.461, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=33.7]
step 9: 97%|█████████▋| 31/32 [00:02<00:00, 19.61it/s, reward=0.382, num_turns=1.94, num_tools=0.935, failed=0, completion_tokens=34.9]
step 9: 100%|██████████| 32/32 [00:02<00:00, 19.61it/s, reward=0.307, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=35.4]
step 9: 100%|██████████| 32/32 [00:02<00:00, 12.05it/s, reward=0.307, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=35.4]
group 0: mean=+0.94 std=0.583 min=+0.5 max=+2.0 | What is the temperature in Tokyo in Fahrenheit?
group 1: mean=+3.92 std=0.220 min=+3.3 max=+4.0 | Convert 22 kg to lbs.
group 2: mean=-1.62 std=1.798 min=-3.0 max=+3.0 | Which is hotter right now, Tokyo or Dubai?
group 3: mean=-2.00 std=0.000 min=-2.0 max=-2.0 | Which is hotter right now, Tokyo or Cairo?
Avg reward: 0.307 | Avg tools/rollout: 0.9 | groups with variance: 3/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0007
Packed 24 trajectories into 2 sequences of length 2048
train: 0%| | 0/2 [00:00<?, ?it/s]The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None}.
==((====))== Unsloth - 2x faster free finetuning | Num GPUs used = 1
\\ /| Num examples = 10,000,000 | Num Epochs = 3 | Total steps = 30,000,000
O^O/ \_/ \ Batch size per device = 2 | Gradient accumulation steps = 1
\ / Data Parallel GPUs = 1 | Total batch size (2 x 1 x 1) = 2
"-____-" Trainable parameters = 4,399,104 of 498,431,872 (0.88% trained)
train: 50%|█████ | 1/2 [00:11<00:11, 11.28s/it]
train: 50%|█████ | 1/2 [00:11<00:11, 11.28s/it, loss/train=-0.343, loss/grad_norm=1.12, loss/learning_rate=5e-5, loss/entropy=0.874]
train: 100%|██████████| 2/2 [00:11<00:00, 4.87s/it, loss/train=-0.343, loss/grad_norm=1.12, loss/learning_rate=5e-5, loss/entropy=0.874]
train: 100%|██████████| 2/2 [00:11<00:00, 4.87s/it, loss/train=-0.0745, loss/grad_norm=1.77, loss/learning_rate=5e-5, loss/entropy=0.299](APIServer pid=14938) Adapters before cleanup: ['default']
(APIServer pid=14938) Keeping active adapter(s): ['default']
(APIServer pid=14938) Adapters after cleanup: ['default']
train: 100%|██████████| 2/2 [00:39<00:00, 19.69s/it, loss/train=-0.0745, loss/grad_norm=1.77, loss/learning_rate=5e-5, loss/entropy=0.299]
============================================================
Step 10/50
============================================================
step 10: 0%| | 0/32 [00:00<?, ?it/s]
step 10: 3%|▎ | 1/32 [00:01<00:43, 1.41s/it]
step 10: 3%|▎ | 1/32 [00:01<00:43, 1.41s/it, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=20.5]
step 10: 6%|▋ | 2/32 [00:01<00:42, 1.41s/it, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=20.5]
step 10: 9%|▉ | 3/32 [00:01<00:40, 1.41s/it, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=20.5]
step 10: 12%|█▎ | 4/32 [00:01<00:39, 1.41s/it, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=20.5]
step 10: 16%|█▌ | 5/32 [00:01<00:38, 1.41s/it, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=20.5]
step 10: 19%|█▉ | 6/32 [00:01<00:36, 1.41s/it, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=20.5]
step 10: 22%|██▏ | 7/32 [00:01<00:35, 1.41s/it, reward=2, num_turns=2, num_tools=1, failed=0, completion_tokens=20.5]
step 10: 25%|██▌ | 8/32 [00:01<00:33, 1.41s/it, reward=1.5, num_turns=2, num_tools=1, failed=0, completion_tokens=21.2]
step 10: 28%|██▊ | 9/32 [00:01<00:32, 1.41s/it, reward=1.11, num_turns=2, num_tools=1, failed=0, completion_tokens=22.2]
step 10: 31%|███▏ | 10/32 [00:01<00:02, 8.57it/s, reward=1.11, num_turns=2, num_tools=1, failed=0, completion_tokens=22.2]
step 10: 31%|███▏ | 10/32 [00:01<00:02, 8.57it/s, reward=0.8, num_turns=2, num_tools=1, failed=0, completion_tokens=22.2]
step 10: 34%|███▍ | 11/32 [00:01<00:02, 8.57it/s, reward=0.545, num_turns=2, num_tools=1, failed=0, completion_tokens=22.6]
step 10: 38%|███▊ | 12/32 [00:01<00:02, 8.57it/s, reward=0.333, num_turns=2, num_tools=1, failed=0, completion_tokens=24]
step 10: 41%|████ | 13/32 [00:01<00:02, 8.57it/s, reward=0.154, num_turns=2, num_tools=1, failed=0, completion_tokens=24.2]
step 10: 44%|████▍ | 14/32 [00:01<00:02, 8.57it/s, reward=0, num_turns=2, num_tools=1, failed=0, completion_tokens=24.5]
step 10: 47%|████▋ | 15/32 [00:01<00:01, 8.57it/s, reward=-0.133, num_turns=2, num_tools=1, failed=0, completion_tokens=24.7]
step 10: 50%|█████ | 16/32 [00:01<00:01, 14.11it/s, reward=-0.133, num_turns=2, num_tools=1, failed=0, completion_tokens=24.7]
step 10: 50%|█████ | 16/32 [00:01<00:01, 14.11it/s, reward=-0.25, num_turns=2, num_tools=1, failed=0, completion_tokens=25.4]
step 10: 53%|█████▎ | 17/32 [00:01<00:01, 14.11it/s, reward=-0.353, num_turns=2, num_tools=1, failed=0, completion_tokens=27]
step 10: 56%|█████▋ | 18/32 [00:01<00:00, 14.11it/s, reward=-0.444, num_turns=2, num_tools=1, failed=0, completion_tokens=27.8]
step 10: 59%|█████▉ | 19/32 [00:01<00:00, 14.11it/s, reward=-0.526, num_turns=2, num_tools=1, failed=0, completion_tokens=29.1]
step 10: 62%|██████▎ | 20/32 [00:01<00:00, 14.11it/s, reward=-0.6, num_turns=2, num_tools=1, failed=0, completion_tokens=29.9]
step 10: 66%|██████▌ | 21/32 [00:01<00:00, 14.11it/s, reward=-0.667, num_turns=2, num_tools=1, failed=0, completion_tokens=30.5]
step 10: 69%|██████▉ | 22/32 [00:01<00:00, 19.98it/s, reward=-0.667, num_turns=2, num_tools=1, failed=0, completion_tokens=30.5]
step 10: 69%|██████▉ | 22/32 [00:01<00:00, 19.98it/s, reward=-0.727, num_turns=2, num_tools=1, failed=0, completion_tokens=31.7]
step 10: 72%|███████▏ | 23/32 [00:01<00:00, 19.98it/s, reward=-0.783, num_turns=2, num_tools=1, failed=0, completion_tokens=33.1]
step 10: 75%|███████▌ | 24/32 [00:01<00:00, 19.98it/s, reward=-0.833, num_turns=2, num_tools=1, failed=0, completion_tokens=34.3]
step 10: 78%|███████▊ | 25/32 [00:01<00:00, 19.98it/s, reward=-0.88, num_turns=2, num_tools=1, failed=0, completion_tokens=35.7]
step 10: 81%|████████▏ | 26/32 [00:01<00:00, 19.98it/s, reward=-0.923, num_turns=2, num_tools=1, failed=0, completion_tokens=36.9]
step 10: 84%|████████▍ | 27/32 [00:01<00:00, 23.30it/s, reward=-0.923, num_turns=2, num_tools=1, failed=0, completion_tokens=36.9]
step 10: 84%|████████▍ | 27/32 [00:01<00:00, 23.30it/s, reward=-0.963, num_turns=2, num_tools=1, failed=0, completion_tokens=38.3]
step 10: 88%|████████▊ | 28/32 [00:01<00:00, 23.30it/s, reward=-1, num_turns=2, num_tools=1, failed=0, completion_tokens=39.1]
step 10: 91%|█████████ | 29/32 [00:01<00:00, 23.30it/s, reward=-1.03, num_turns=2, num_tools=1, failed=0, completion_tokens=40.6]
step 10: 94%|█████████▍| 30/32 [00:01<00:00, 23.30it/s, reward=-1.07, num_turns=2, num_tools=1, failed=0, completion_tokens=40]
step 10: 97%|█████████▋| 31/32 [00:02<00:00, 23.30it/s, reward=-1.1, num_turns=2.03, num_tools=1.03, failed=0, completion_tokens=40.3]
step 10: 100%|██████████| 32/32 [00:03<00:00, 7.29it/s, reward=-1.1, num_turns=2.03, num_tools=1.03, failed=0, completion_tokens=40.3]
step 10: 100%|██████████| 32/32 [00:03<00:00, 7.29it/s, reward=-1.12, num_turns=2.03, num_tools=1.03, failed=0, completion_tokens=47.7]
step 10: 100%|██████████| 32/32 [00:03<00:00, 8.82it/s, reward=-1.12, num_turns=2.03, num_tools=1.03, failed=0, completion_tokens=47.7]
group 0: mean=-2.00 std=0.000 min=-2.0 max=-2.0 | What is the speed of light?
group 1: mean=-2.00 std=0.000 min=-2.0 max=-2.0 | What is the GDP of India?
group 2: mean=-2.00 std=0.000 min=-2.0 max=-2.0 | What is the GDP of Germany?
group 3: mean=+1.50 std=1.323 min=-2.0 max=+2.0 | What is 404 minus 5?
Avg reward: -1.125 | Avg tools/rollout: 1.0 | groups with variance: 1/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0008
Packed 8 trajectories into 1 sequences of length 2048
train: 0%| | 0/1 [00:00<?, ?it/s]
train: 100%|██████████| 1/1 [00:01<00:00, 1.87s/it]
train: 100%|██████████| 1/1 [00:01<00:00, 1.87s/it, loss/train=-0.524, loss/grad_norm=2.19, loss/learning_rate=5e-5, loss/entropy=0.304](APIServer pid=14938) Adapters before cleanup: ['default']
(APIServer pid=14938) Keeping active adapter(s): ['default']
(APIServer pid=14938) Adapters after cleanup: ['default']
train: 100%|██████████| 1/1 [00:29<00:00, 29.74s/it, loss/train=-0.524, loss/grad_norm=2.19, loss/learning_rate=5e-5, loss/entropy=0.304]
Running validation...
validation: 0%| | 0/400 [00:00<?, ?it/s]
validation: 0%| | 1/400 [00:13<1:30:20, 13.59s/it]
validation: 0%| | 1/400 [00:13<1:30:20, 13.59s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=37]
validation: 0%| | 2/400 [00:13<37:45, 5.69s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=37]
validation: 0%| | 2/400 [00:13<37:45, 5.69s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=52]
validation: 1%| | 3/400 [00:13<20:50, 3.15s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=52]
validation: 1%| | 3/400 [00:13<20:50, 3.15s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=36.3]
validation: 1%| | 4/400 [00:13<20:46, 3.15s/it, reward=-1.25, num_turns=1.25, num_tools=0.25, failed=0, completion_tokens=34.2]
validation: 1%|▏ | 5/400 [00:13<20:43, 3.15s/it, reward=-0.2, num_turns=1.4, num_tools=0.4, failed=0, completion_tokens=33]
validation: 2%|▏ | 6/400 [00:13<20:40, 3.15s/it, reward=0.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=32.3]
validation: 2%|▏ | 7/400 [00:13<20:37, 3.15s/it, reward=1, num_turns=1.57, num_tools=0.571, failed=0, completion_tokens=31.9]
validation: 2%|▏ | 8/400 [00:13<20:34, 3.15s/it, reward=1.12, num_turns=1.62, num_tools=0.625, failed=0, completion_tokens=31.4]
validation: 2%|▏ | 9/400 [00:13<20:31, 3.15s/it, reward=1.22, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=31.6]
validation: 2%|▎ | 10/400 [00:13<20:28, 3.15s/it, reward=1.25, num_turns=1.7, num_tools=0.7, failed=0, completion_tokens=32.4]
validation: 3%|▎ | 11/400 [00:13<20:24, 3.15s/it, reward=0.955, num_turns=1.73, num_tools=0.727, failed=0, completion_tokens=32.4]
validation: 3%|▎ | 12/400 [00:13<20:21, 3.15s/it, reward=0.708, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=32]
validation: 3%|▎ | 13/400 [00:13<20:18, 3.15s/it, reward=0.5, num_turns=1.77, num_tools=0.769, failed=0, completion_tokens=32.2]
validation: 4%|▎ | 14/400 [00:13<20:15, 3.15s/it, reward=0.25, num_turns=1.71, num_tools=0.714, failed=0, completion_tokens=33.6]
validation: 4%|▍ | 15/400 [00:13<20:12, 3.15s/it, reward=0.0333, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=34.8]
validation: 4%|▍ | 16/400 [00:14<02:27, 2.61it/s, reward=0.0333, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=34.8]
validation: 4%|▍ | 16/400 [00:14<02:27, 2.61it/s, reward=-0.156, num_turns=1.62, num_tools=0.625, failed=0, completion_tokens=34.8]
validation: 4%|▍ | 17/400 [00:14<02:26, 2.61it/s, reward=-0.324, num_turns=1.59, num_tools=0.588, failed=0, completion_tokens=39.1]
validation: 4%|▍ | 18/400 [00:15<02:33, 2.49it/s, reward=-0.324, num_turns=1.59, num_tools=0.588, failed=0, completion_tokens=39.1]
validation: 4%|▍ | 18/400 [00:15<02:33, 2.49it/s, reward=-0.417, num_turns=1.61, num_tools=0.611, failed=0, completion_tokens=40.5]
validation: 5%|▍ | 19/400 [00:15<02:32, 2.49it/s, reward=-0.553, num_turns=1.58, num_tools=0.579, failed=0, completion_tokens=41.1]
validation: 5%|▌ | 20/400 [00:15<02:32, 2.49it/s, reward=-0.675, num_turns=1.55, num_tools=0.55, failed=0, completion_tokens=46.6]
validation: 5%|▌ | 21/400 [00:15<02:32, 2.49it/s, reward=-0.548, num_turns=1.57, num_tools=0.571, failed=0, completion_tokens=45.6]
validation: 6%|▌ | 22/400 [00:15<02:31, 2.49it/s, reward=-0.432, num_turns=1.59, num_tools=0.591, failed=0, completion_tokens=44.8]
validation: 6%|▌ | 23/400 [00:15<02:31, 2.49it/s, reward=-0.326, num_turns=1.61, num_tools=0.609, failed=0, completion_tokens=44.1]
validation: 6%|▌ | 24/400 [00:15<02:30, 2.49it/s, reward=-0.292, num_turns=1.62, num_tools=0.625, failed=0, completion_tokens=43.4]
validation: 6%|▋ | 25/400 [00:15<02:30, 2.49it/s, reward=-0.26, num_turns=1.64, num_tools=0.64, failed=0, completion_tokens=42.8]
validation: 6%|▋ | 26/400 [00:15<02:30, 2.49it/s, reward=-0.327, num_turns=1.65, num_tools=0.654, failed=0, completion_tokens=41.9]
validation: 7%|▋ | 27/400 [00:15<02:29, 2.49it/s, reward=-0.389, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=41.3]
validation: 7%|▋ | 28/400 [00:15<02:29, 2.49it/s, reward=-0.446, num_turns=1.68, num_tools=0.679, failed=0, completion_tokens=40.8]
validation: 7%|▋ | 29/400 [00:15<02:28, 2.49it/s, reward=-0.293, num_turns=1.69, num_tools=0.69, failed=0, completion_tokens=40.2]
validation: 8%|▊ | 30/400 [00:15<01:04, 5.75it/s, reward=-0.293, num_turns=1.69, num_tools=0.69, failed=0, completion_tokens=40.2]
validation: 8%|▊ | 30/400 [00:15<01:04, 5.75it/s, reward=-0.35, num_turns=1.7, num_tools=0.7, failed=0, completion_tokens=40.3]
validation: 8%|▊ | 31/400 [00:15<01:04, 5.75it/s, reward=-0.403, num_turns=1.71, num_tools=0.742, failed=0, completion_tokens=39.9]
validation: 8%|▊ | 32/400 [00:15<01:04, 5.75it/s, reward=-0.453, num_turns=1.72, num_tools=0.75, failed=0, completion_tokens=39.5]
validation: 8%|▊ | 33/400 [00:15<01:03, 5.75it/s, reward=-0.5, num_turns=1.73, num_tools=0.758, failed=0, completion_tokens=39]
validation: 8%|▊ | 34/400 [00:15<00:53, 6.84it/s, reward=-0.5, num_turns=1.73, num_tools=0.758, failed=0, completion_tokens=39]
validation: 8%|▊ | 34/400 [00:15<00:53, 6.84it/s, reward=-0.544, num_turns=1.74, num_tools=0.765, failed=0, completion_tokens=38.6]
validation: 9%|▉ | 35/400 [00:15<00:53, 6.84it/s, reward=-0.586, num_turns=1.74, num_tools=0.771, failed=0, completion_tokens=38.4]
validation: 9%|▉ | 36/400 [00:15<00:53, 6.84it/s, reward=-0.625, num_turns=1.75, num_tools=0.778, failed=0, completion_tokens=38.2]
validation: 9%|▉ | 37/400 [00:15<00:53, 6.84it/s, reward=-0.689, num_turns=1.73, num_tools=0.757, failed=0, completion_tokens=44.9]
validation: 10%|▉ | 38/400 [00:15<00:52, 6.84it/s, reward=-0.583, num_turns=1.74, num_tools=0.763, failed=0, completion_tokens=44.5]
validation: 10%|▉ | 39/400 [00:15<00:52, 6.84it/s, reward=-0.62, num_turns=1.74, num_tools=0.769, failed=0, completion_tokens=44.2]
validation: 10%|█ | 40/400 [00:15<00:52, 6.84it/s, reward=-0.679, num_turns=1.73, num_tools=0.75, failed=0, completion_tokens=48.1]
validation: 10%|█ | 41/400 [00:16<00:36, 9.91it/s, reward=-0.679, num_turns=1.73, num_tools=0.75, failed=0, completion_tokens=48.1]
validation: 10%|█ | 41/400 [00:16<00:36, 9.91it/s, reward=-0.614, num_turns=1.73, num_tools=0.756, failed=0, completion_tokens=47.6]
validation: 10%|█ | 42/400 [00:16<00:36, 9.91it/s, reward=-0.647, num_turns=1.74, num_tools=0.762, failed=0, completion_tokens=47.1]
validation: 11%|█ | 43/400 [00:16<00:36, 9.91it/s, reward=-0.539, num_turns=1.74, num_tools=0.767, failed=0, completion_tokens=46.7]
validation: 11%|█ | 44/400 [00:16<00:35, 9.91it/s, reward=-0.436, num_turns=1.75, num_tools=0.773, failed=0, completion_tokens=46.2]
validation: 11%|█▏ | 45/400 [00:16<00:35, 9.91it/s, reward=-0.337, num_turns=1.76, num_tools=0.778, failed=0, completion_tokens=45.9]
validation: 12%|█▏ | 46/400 [00:16<00:35, 9.91it/s, reward=-0.243, num_turns=1.76, num_tools=0.783, failed=0, completion_tokens=45.5]
validation: 12%|█▏ | 47/400 [00:16<00:35, 9.91it/s, reward=-0.152, num_turns=1.77, num_tools=0.787, failed=0, completion_tokens=45.1]
validation: 12%|█▏ | 48/400 [00:16<00:35, 9.91it/s, reward=-0.128, num_turns=1.77, num_tools=0.792, failed=0, completion_tokens=44.9]
validation: 12%|█▏ | 49/400 [00:16<00:35, 9.91it/s, reward=-0.105, num_turns=1.78, num_tools=0.796, failed=0, completion_tokens=44.6]
validation: 12%|█▎ | 50/400 [00:16<00:35, 9.91it/s, reward=-0.143, num_turns=1.78, num_tools=0.8, failed=0, completion_tokens=44.4]
validation: 13%|█▎ | 51/400 [00:16<00:35, 9.91it/s, reward=-0.121, num_turns=1.78, num_tools=0.804, failed=0, completion_tokens=44.6]
validation: 13%|█▎ | 52/400 [00:16<00:35, 9.91it/s, reward=-0.176, num_turns=1.77, num_tools=0.788, failed=0, completion_tokens=50.6]
validation: 13%|█▎ | 53/400 [00:16<00:35, 9.91it/s, reward=-0.16, num_turns=1.77, num_tools=0.792, failed=0, completion_tokens=50.5]
validation: 14%|█▎ | 54/400 [00:16<00:34, 9.91it/s, reward=-0.194, num_turns=1.78, num_tools=0.796, failed=0, completion_tokens=50.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 14%|█▍ | 55/400 [00:16<00:18, 18.42it/s, reward=-0.194, num_turns=1.78, num_tools=0.796, failed=0, completion_tokens=50.1]
validation: 14%|█▍ | 55/400 [00:16<00:18, 18.42it/s, reward=-0.158, num_turns=1.78, num_tools=0.8, failed=0, completion_tokens=49.8]
validation: 14%|█▍ | 56/400 [00:16<00:18, 18.42it/s, reward=-0.119, num_turns=1.79, num_tools=0.804, failed=0, completion_tokens=49.5][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 14%|█▍ | 57/400 [00:16<00:18, 18.42it/s, reward=-0.17, num_turns=1.77, num_tools=0.789, failed=0.0175, completion_tokens=49.5]
validation: 14%|█▍ | 58/400 [00:16<00:18, 18.42it/s, reward=-0.218, num_turns=1.76, num_tools=0.776, failed=0.0345, completion_tokens=49.5]
validation: 15%|█▍ | 59/400 [00:16<00:18, 18.42it/s, reward=-0.266, num_turns=1.75, num_tools=0.763, failed=0.0508, completion_tokens=49.5]
validation: 15%|█▌ | 60/400 [00:16<00:18, 18.42it/s, reward=-0.311, num_turns=1.73, num_tools=0.75, failed=0.0667, completion_tokens=49.5]
validation: 15%|█▌ | 61/400 [00:16<00:18, 18.42it/s, reward=-0.355, num_turns=1.72, num_tools=0.738, failed=0.082, completion_tokens=49.5]
validation: 16%|█▌ | 62/400 [00:16<00:18, 18.42it/s, reward=-0.398, num_turns=1.71, num_tools=0.726, failed=0.0968, completion_tokens=49.5]
validation: 16%|█▌ | 63/400 [00:16<00:18, 18.42it/s, reward=-0.439, num_turns=1.7, num_tools=0.714, failed=0.111, completion_tokens=49.5]
validation: 16%|█▌ | 64/400 [00:16<00:18, 18.42it/s, reward=-0.479, num_turns=1.69, num_tools=0.703, failed=0.125, completion_tokens=49.5]
validation: 16%|█▋ | 65/400 [00:16<00:18, 18.42it/s, reward=-0.518, num_turns=1.68, num_tools=0.692, failed=0.138, completion_tokens=49.5]
validation: 16%|█▋ | 66/400 [00:16<00:18, 18.42it/s, reward=-0.556, num_turns=1.67, num_tools=0.682, failed=0.152, completion_tokens=49.5]
validation: 17%|█▋ | 67/400 [00:16<00:18, 18.42it/s, reward=-0.592, num_turns=1.66, num_tools=0.672, failed=0.164, completion_tokens=49.5]
validation: 17%|█▋ | 68/400 [00:16<00:18, 18.42it/s, reward=-0.627, num_turns=1.65, num_tools=0.662, failed=0.176, completion_tokens=49.5]
validation: 17%|█▋ | 69/400 [00:16<00:17, 18.42it/s, reward=-0.662, num_turns=1.64, num_tools=0.652, failed=0.188, completion_tokens=49.5]
validation: 18%|█▊ | 70/400 [00:16<00:17, 18.42it/s, reward=-0.695, num_turns=1.63, num_tools=0.643, failed=0.2, completion_tokens=49.5]
validation: 18%|█▊ | 71/400 [00:16<00:17, 18.42it/s, reward=-0.728, num_turns=1.62, num_tools=0.634, failed=0.211, completion_tokens=49.5]
validation: 18%|█▊ | 72/400 [00:16<00:17, 18.42it/s, reward=-0.759, num_turns=1.61, num_tools=0.625, failed=0.222, completion_tokens=49.5]
validation: 18%|█▊ | 73/400 [00:16<00:17, 18.42it/s, reward=-0.79, num_turns=1.6, num_tools=0.616, failed=0.233, completion_tokens=49.5]
validation: 18%|█▊ | 74/400 [00:16<00:17, 18.42it/s, reward=-0.82, num_turns=1.59, num_tools=0.608, failed=0.243, completion_tokens=49.5]
validation: 19%|█▉ | 75/400 [00:16<00:17, 18.42it/s, reward=-0.849, num_turns=1.59, num_tools=0.6, failed=0.253, completion_tokens=49.5]
validation: 19%|█▉ | 76/400 [00:16<00:17, 18.42it/s, reward=-0.877, num_turns=1.58, num_tools=0.592, failed=0.263, completion_tokens=49.5]
validation: 19%|█▉ | 77/400 [00:16<00:17, 18.42it/s, reward=-0.905, num_turns=1.57, num_tools=0.584, failed=0.273, completion_tokens=49.5]
validation: 20%|█▉ | 78/400 [00:16<00:17, 18.42it/s, reward=-0.932, num_turns=1.56, num_tools=0.577, failed=0.282, completion_tokens=49.5]
validation: 20%|█▉ | 79/400 [00:16<00:17, 18.42it/s, reward=-0.958, num_turns=1.56, num_tools=0.57, failed=0.291, completion_tokens=49.5]
validation: 20%|██ | 80/400 [00:16<00:17, 18.42it/s, reward=-0.983, num_turns=1.55, num_tools=0.562, failed=0.3, completion_tokens=49.5]
validation: 20%|██ | 81/400 [00:16<00:17, 18.42it/s, reward=-1.01, num_turns=1.54, num_tools=0.556, failed=0.309, completion_tokens=49.5]
validation: 20%|██ | 82/400 [00:16<00:17, 18.42it/s, reward=-1.03, num_turns=1.54, num_tools=0.549, failed=0.317, completion_tokens=49.5]
validation: 21%|██ | 83/400 [00:16<00:17, 18.42it/s, reward=-1.06, num_turns=1.53, num_tools=0.542, failed=0.325, completion_tokens=49.5]
validation: 21%|██ | 84/400 [00:16<00:17, 18.42it/s, reward=-1.08, num_turns=1.52, num_tools=0.536, failed=0.333, completion_tokens=49.5]
validation: 21%|██▏ | 85/400 [00:16<00:17, 18.42it/s, reward=-1.1, num_turns=1.52, num_tools=0.529, failed=0.341, completion_tokens=49.5]
validation: 22%|██▏ | 86/400 [00:16<00:17, 18.42it/s, reward=-1.12, num_turns=1.51, num_tools=0.523, failed=0.349, completion_tokens=49.5]
validation: 22%|██▏ | 87/400 [00:16<00:16, 18.42it/s, reward=-1.15, num_turns=1.51, num_tools=0.517, failed=0.356, completion_tokens=49.5]
validation: 22%|██▏ | 88/400 [00:16<00:16, 18.42it/s, reward=-1.17, num_turns=1.5, num_tools=0.511, failed=0.364, completion_tokens=49.5]
validation: 22%|██▏ | 89/400 [00:16<00:16, 18.42it/s, reward=-1.19, num_turns=1.49, num_tools=0.506, failed=0.371, completion_tokens=49.5]
validation: 22%|██▎ | 90/400 [00:16<00:16, 18.42it/s, reward=-1.2, num_turns=1.5, num_tools=0.511, failed=0.367, completion_tokens=49.2] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 23%|██▎ | 91/400 [00:16<00:16, 18.42it/s, reward=-1.21, num_turns=1.51, num_tools=0.516, failed=0.363, completion_tokens=49.6][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 23%|██▎ | 92/400 [00:16<00:16, 18.42it/s, reward=-1.22, num_turns=1.5, num_tools=0.511, failed=0.37, completion_tokens=49.6]
validation: 23%|██▎ | 93/400 [00:16<00:16, 18.42it/s, reward=-1.24, num_turns=1.49, num_tools=0.505, failed=0.376, completion_tokens=49.6]
validation: 24%|██▎ | 94/400 [00:16<00:16, 18.42it/s, reward=-1.26, num_turns=1.49, num_tools=0.5, failed=0.383, completion_tokens=49.6]
validation: 24%|██▍ | 95/400 [00:16<00:06, 50.79it/s, reward=-1.26, num_turns=1.49, num_tools=0.5, failed=0.383, completion_tokens=49.6]
validation: 24%|██▍ | 95/400 [00:16<00:06, 50.79it/s, reward=-1.28, num_turns=1.48, num_tools=0.495, failed=0.389, completion_tokens=49.6]
validation: 24%|██▍ | 96/400 [00:16<00:05, 50.79it/s, reward=-1.3, num_turns=1.48, num_tools=0.49, failed=0.396, completion_tokens=49.6] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 24%|██▍ | 97/400 [00:16<00:05, 50.79it/s, reward=-1.32, num_turns=1.47, num_tools=0.485, failed=0.402, completion_tokens=49.6]
validation: 24%|██▍ | 98/400 [00:16<00:05, 50.79it/s, reward=-1.33, num_turns=1.47, num_tools=0.48, failed=0.408, completion_tokens=49.6]
validation: 25%|██▍ | 99/400 [00:16<00:05, 50.79it/s, reward=-1.35, num_turns=1.46, num_tools=0.475, failed=0.414, completion_tokens=49.6][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 25%|██▌ | 100/400 [00:16<00:05, 50.79it/s, reward=-1.37, num_turns=1.46, num_tools=0.47, failed=0.42, completion_tokens=49.6]
validation: 25%|██▌ | 101/400 [00:16<00:05, 50.79it/s, reward=-1.38, num_turns=1.46, num_tools=0.465, failed=0.426, completion_tokens=49.6]
validation: 26%|██▌ | 102/400 [00:16<00:05, 50.79it/s, reward=-1.4, num_turns=1.45, num_tools=0.461, failed=0.431, completion_tokens=49.6]
validation: 26%|██▌ | 103/400 [00:16<00:05, 50.79it/s, reward=-1.41, num_turns=1.45, num_tools=0.456, failed=0.437, completion_tokens=49.6]
validation: 26%|██▌ | 104/400 [00:16<00:05, 50.79it/s, reward=-1.43, num_turns=1.44, num_tools=0.452, failed=0.442, completion_tokens=49.6]
validation: 26%|██▋ | 105/400 [00:16<00:05, 50.79it/s, reward=-1.44, num_turns=1.44, num_tools=0.448, failed=0.448, completion_tokens=49.6]
validation: 26%|██▋ | 106/400 [00:16<00:05, 50.79it/s, reward=-1.46, num_turns=1.43, num_tools=0.443, failed=0.453, completion_tokens=49.6]
validation: 27%|██▋ | 107/400 [00:16<00:05, 50.79it/s, reward=-1.47, num_turns=1.43, num_tools=0.439, failed=0.458, completion_tokens=49.6]
validation: 27%|██▋ | 108/400 [00:16<00:05, 50.79it/s, reward=-1.49, num_turns=1.43, num_tools=0.435, failed=0.463, completion_tokens=49.6]
validation: 27%|██▋ | 109/400 [00:16<00:05, 50.79it/s, reward=-1.5, num_turns=1.42, num_tools=0.431, failed=0.468, completion_tokens=49.6]
validation: 28%|██▊ | 110/400 [00:16<00:05, 50.79it/s, reward=-1.52, num_turns=1.42, num_tools=0.427, failed=0.473, completion_tokens=49.6]
validation: 28%|██▊ | 111/400 [00:16<00:06, 42.54it/s, reward=-1.52, num_turns=1.42, num_tools=0.427, failed=0.473, completion_tokens=49.6]
validation: 28%|██▊ | 111/400 [00:16<00:06, 42.54it/s, reward=-1.53, num_turns=1.41, num_tools=0.423, failed=0.477, completion_tokens=49.6]
validation: 28%|██▊ | 112/400 [00:16<00:06, 42.54it/s, reward=-1.54, num_turns=1.41, num_tools=0.42, failed=0.482, completion_tokens=49.6]
validation: 28%|██▊ | 113/400 [00:16<00:06, 42.54it/s, reward=-1.55, num_turns=1.41, num_tools=0.416, failed=0.487, completion_tokens=49.6]
validation: 28%|██▊ | 114/400 [00:16<00:06, 42.54it/s, reward=-1.57, num_turns=1.4, num_tools=0.412, failed=0.491, completion_tokens=49.6]
validation: 29%|██▉ | 115/400 [00:16<00:06, 42.54it/s, reward=-1.58, num_turns=1.4, num_tools=0.409, failed=0.496, completion_tokens=49.6]
validation: 29%|██▉ | 116/400 [00:16<00:06, 42.54it/s, reward=-1.59, num_turns=1.4, num_tools=0.405, failed=0.5, completion_tokens=49.6]
validation: 29%|██▉ | 117/400 [00:16<00:06, 42.54it/s, reward=-1.6, num_turns=1.39, num_tools=0.402, failed=0.504, completion_tokens=49.6]
validation: 30%|██▉ | 118/400 [00:16<00:06, 42.54it/s, reward=-1.62, num_turns=1.39, num_tools=0.398, failed=0.508, completion_tokens=49.6]
validation: 30%|██▉ | 119/400 [00:16<00:06, 42.54it/s, reward=-1.63, num_turns=1.39, num_tools=0.395, failed=0.513, completion_tokens=49.6]
validation: 30%|███ | 120/400 [00:16<00:06, 42.54it/s, reward=-1.64, num_turns=1.38, num_tools=0.392, failed=0.517, completion_tokens=49.6]
validation: 30%|███ | 121/400 [00:16<00:06, 42.54it/s, reward=-1.65, num_turns=1.38, num_tools=0.388, failed=0.521, completion_tokens=49.6]
validation: 30%|███ | 122/400 [00:16<00:06, 42.54it/s, reward=-1.66, num_turns=1.38, num_tools=0.385, failed=0.525, completion_tokens=49.6]
validation: 31%|███ | 123/400 [00:16<00:06, 42.54it/s, reward=-1.67, num_turns=1.37, num_tools=0.382, failed=0.528, completion_tokens=49.6]
validation: 31%|███ | 124/400 [00:16<00:06, 42.54it/s, reward=-1.68, num_turns=1.37, num_tools=0.379, failed=0.532, completion_tokens=49.6][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 31%|███▏ | 125/400 [00:17<00:05, 46.65it/s, reward=-1.68, num_turns=1.37, num_tools=0.379, failed=0.532, completion_tokens=49.6]
validation: 31%|███▏ | 125/400 [00:17<00:05, 46.65it/s, reward=-1.69, num_turns=1.37, num_tools=0.376, failed=0.536, completion_tokens=49.6]
validation: 32%|███▏ | 126/400 [00:17<00:05, 46.65it/s, reward=-1.7, num_turns=1.37, num_tools=0.373, failed=0.54, completion_tokens=49.6]
validation: 32%|███▏ | 127/400 [00:17<00:05, 46.65it/s, reward=-1.71, num_turns=1.36, num_tools=0.37, failed=0.543, completion_tokens=49.6]
validation: 32%|███▏ | 128/400 [00:17<00:05, 46.65it/s, reward=-1.72, num_turns=1.36, num_tools=0.367, failed=0.547, completion_tokens=49.6]
validation: 32%|███▏ | 129/400 [00:17<00:05, 46.65it/s, reward=-1.73, num_turns=1.36, num_tools=0.364, failed=0.55, completion_tokens=49.6]
validation: 32%|███▎ | 130/400 [00:17<00:05, 46.65it/s, reward=-1.74, num_turns=1.35, num_tools=0.362, failed=0.554, completion_tokens=49.6]
validation: 33%|███▎ | 131/400 [00:17<00:05, 46.65it/s, reward=-1.75, num_turns=1.35, num_tools=0.359, failed=0.557, completion_tokens=49.6]
validation: 33%|███▎ | 132/400 [00:17<00:05, 46.65it/s, reward=-1.76, num_turns=1.35, num_tools=0.356, failed=0.561, completion_tokens=49.6]
validation: 33%|███▎ | 133/400 [00:17<00:05, 46.65it/s, reward=-1.77, num_turns=1.35, num_tools=0.353, failed=0.564, completion_tokens=49.6]
validation: 34%|███▎ | 134/400 [00:17<00:05, 46.65it/s, reward=-1.74, num_turns=1.35, num_tools=0.358, failed=0.56, completion_tokens=49.3]
validation: 34%|███▍ | 135/400 [00:17<00:05, 46.65it/s, reward=-1.75, num_turns=1.35, num_tools=0.356, failed=0.556, completion_tokens=49.3]
validation: 34%|███▍ | 136/400 [00:17<00:05, 46.65it/s, reward=-1.76, num_turns=1.35, num_tools=0.353, failed=0.551, completion_tokens=49.4]
validation: 34%|███▍ | 137/400 [00:17<00:05, 46.65it/s, reward=-1.74, num_turns=1.35, num_tools=0.358, failed=0.547, completion_tokens=49.2]
validation: 34%|███▍ | 138/400 [00:17<00:05, 46.65it/s, reward=-1.74, num_turns=1.36, num_tools=0.362, failed=0.543, completion_tokens=49]
validation: 35%|███▍ | 139/400 [00:17<00:05, 46.65it/s, reward=-1.75, num_turns=1.35, num_tools=0.36, failed=0.547, completion_tokens=49]
validation: 35%|███▌ | 140/400 [00:17<00:05, 46.65it/s, reward=-1.76, num_turns=1.35, num_tools=0.357, failed=0.55, completion_tokens=49]
validation: 35%|███▌ | 141/400 [00:17<00:05, 46.65it/s, reward=-1.76, num_turns=1.35, num_tools=0.355, failed=0.553, completion_tokens=49]
validation: 36%|███▌ | 142/400 [00:17<00:05, 46.65it/s, reward=-1.77, num_turns=1.35, num_tools=0.352, failed=0.556, completion_tokens=49]
validation: 36%|███▌ | 143/400 [00:17<00:05, 46.65it/s, reward=-1.78, num_turns=1.34, num_tools=0.35, failed=0.559, completion_tokens=49]
validation: 36%|███▌ | 144/400 [00:17<00:05, 46.65it/s, reward=-1.79, num_turns=1.34, num_tools=0.347, failed=0.562, completion_tokens=49]
validation: 36%|███▋ | 145/400 [00:17<00:05, 46.65it/s, reward=-1.8, num_turns=1.34, num_tools=0.345, failed=0.566, completion_tokens=49]
validation: 36%|███▋ | 146/400 [00:17<00:05, 46.65it/s, reward=-1.81, num_turns=1.34, num_tools=0.342, failed=0.568, completion_tokens=49]
validation: 37%|███▋ | 147/400 [00:17<00:05, 46.65it/s, reward=-1.82, num_turns=1.33, num_tools=0.34, failed=0.571, completion_tokens=49]
validation: 37%|███▋ | 148/400 [00:17<00:05, 46.65it/s, reward=-1.82, num_turns=1.33, num_tools=0.338, failed=0.574, completion_tokens=49]
validation: 37%|███▋ | 149/400 [00:17<00:05, 46.65it/s, reward=-1.83, num_turns=1.33, num_tools=0.336, failed=0.577, completion_tokens=49]
validation: 38%|███▊ | 150/400 [00:17<00:05, 46.65it/s, reward=-1.84, num_turns=1.33, num_tools=0.333, failed=0.58, completion_tokens=49]
validation: 38%|███▊ | 151/400 [00:17<00:05, 46.65it/s, reward=-1.85, num_turns=1.32, num_tools=0.331, failed=0.583, completion_tokens=49]
validation: 38%|███▊ | 152/400 [00:17<00:05, 46.65it/s, reward=-1.85, num_turns=1.32, num_tools=0.329, failed=0.586, completion_tokens=49]
validation: 38%|███▊ | 153/400 [00:17<00:05, 46.65it/s, reward=-1.86, num_turns=1.32, num_tools=0.327, failed=0.588, completion_tokens=49]
validation: 38%|███▊ | 154/400 [00:17<00:05, 46.65it/s, reward=-1.87, num_turns=1.32, num_tools=0.325, failed=0.591, completion_tokens=49]
validation: 39%|███▉ | 155/400 [00:17<00:05, 46.65it/s, reward=-1.88, num_turns=1.32, num_tools=0.323, failed=0.594, completion_tokens=49]
validation: 39%|███▉ | 156/400 [00:17<00:05, 46.65it/s, reward=-1.88, num_turns=1.31, num_tools=0.321, failed=0.596, completion_tokens=49]
validation: 39%|███▉ | 157/400 [00:17<00:05, 46.65it/s, reward=-1.89, num_turns=1.31, num_tools=0.318, failed=0.599, completion_tokens=49]
validation: 40%|███▉ | 158/400 [00:17<00:05, 46.65it/s, reward=-1.9, num_turns=1.31, num_tools=0.316, failed=0.601, completion_tokens=49]
validation: 40%|███▉ | 159/400 [00:17<00:05, 46.65it/s, reward=-1.9, num_turns=1.31, num_tools=0.314, failed=0.604, completion_tokens=49]
validation: 40%|████ | 160/400 [00:17<00:05, 46.65it/s, reward=-1.91, num_turns=1.31, num_tools=0.312, failed=0.606, completion_tokens=49]
validation: 40%|████ | 161/400 [00:17<00:05, 46.65it/s, reward=-1.92, num_turns=1.3, num_tools=0.311, failed=0.609, completion_tokens=49]
validation: 40%|████ | 162/400 [00:17<00:05, 46.65it/s, reward=-1.92, num_turns=1.3, num_tools=0.309, failed=0.611, completion_tokens=49]
validation: 41%|████ | 163/400 [00:17<00:05, 46.65it/s, reward=-1.93, num_turns=1.3, num_tools=0.307, failed=0.613, completion_tokens=49]
validation: 41%|████ | 164/400 [00:17<00:05, 46.65it/s, reward=-1.94, num_turns=1.3, num_tools=0.305, failed=0.616, completion_tokens=49]
validation: 41%|████▏ | 165/400 [00:17<00:05, 46.65it/s, reward=-1.94, num_turns=1.3, num_tools=0.303, failed=0.618, completion_tokens=49]
validation: 42%|████▏ | 166/400 [00:17<00:05, 46.65it/s, reward=-1.95, num_turns=1.3, num_tools=0.301, failed=0.62, completion_tokens=49]
validation: 42%|████▏ | 167/400 [00:17<00:04, 46.65it/s, reward=-1.96, num_turns=1.29, num_tools=0.299, failed=0.623, completion_tokens=49]
validation: 42%|████▏ | 168/400 [00:17<00:04, 46.65it/s, reward=-1.96, num_turns=1.29, num_tools=0.298, failed=0.625, completion_tokens=49]
validation: 42%|████▏ | 169/400 [00:17<00:04, 46.65it/s, reward=-1.97, num_turns=1.29, num_tools=0.296, failed=0.621, completion_tokens=50.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 42%|████▎ | 170/400 [00:17<00:04, 46.65it/s, reward=-1.98, num_turns=1.29, num_tools=0.294, failed=0.624, completion_tokens=50.8]
validation: 43%|████▎ | 171/400 [00:17<00:04, 46.65it/s, reward=-1.98, num_turns=1.29, num_tools=0.292, failed=0.626, completion_tokens=50.8]
validation: 43%|████▎ | 172/400 [00:17<00:04, 46.65it/s, reward=-1.99, num_turns=1.29, num_tools=0.297, failed=0.628, completion_tokens=50.6]
validation: 43%|████▎ | 173/400 [00:17<00:04, 46.65it/s, reward=-1.99, num_turns=1.29, num_tools=0.301, failed=0.63, completion_tokens=50.4]
validation: 44%|████▎ | 174/400 [00:17<00:04, 46.65it/s, reward=-2, num_turns=1.3, num_tools=0.305, failed=0.632, completion_tokens=50.1]
validation: 44%|████▍ | 175/400 [00:17<00:04, 46.65it/s, reward=-2, num_turns=1.3, num_tools=0.309, failed=0.634, completion_tokens=49.9]
validation: 44%|████▍ | 176/400 [00:17<00:04, 46.65it/s, reward=-2.01, num_turns=1.31, num_tools=0.312, failed=0.636, completion_tokens=49.7]
validation: 44%|████▍ | 177/400 [00:17<00:04, 46.65it/s, reward=-2.02, num_turns=1.31, num_tools=0.316, failed=0.638, completion_tokens=49.3]
validation: 44%|████▍ | 178/400 [00:17<00:04, 46.65it/s, reward=-2.02, num_turns=1.31, num_tools=0.315, failed=0.64, completion_tokens=49.3]
validation: 45%|████▍ | 179/400 [00:17<00:04, 46.65it/s, reward=-2.03, num_turns=1.31, num_tools=0.318, failed=0.642, completion_tokens=48.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 45%|████▌ | 180/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.31, num_tools=0.318, failed=0.642, completion_tokens=48.9]
validation: 45%|████▌ | 180/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.31, num_tools=0.317, failed=0.644, completion_tokens=48.9]
validation: 45%|████▌ | 181/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.31, num_tools=0.32, failed=0.641, completion_tokens=48.6]
validation: 46%|████▌ | 182/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.32, num_tools=0.324, failed=0.637, completion_tokens=48.3]
validation: 46%|████▌ | 183/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.32, num_tools=0.328, failed=0.634, completion_tokens=48]
validation: 46%|████▌ | 184/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.33, num_tools=0.332, failed=0.63, completion_tokens=47.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 46%|████▋ | 185/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.33, num_tools=0.335, failed=0.627, completion_tokens=47.4]
validation: 46%|████▋ | 186/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.33, num_tools=0.339, failed=0.624, completion_tokens=47.1]
validation: 47%|████▋ | 187/400 [00:17<00:02, 93.58it/s, reward=-2.03, num_turns=1.34, num_tools=0.342, failed=0.62, completion_tokens=46.7]
validation: 47%|████▋ | 188/400 [00:17<00:02, 93.58it/s, reward=-2, num_turns=1.34, num_tools=0.346, failed=0.617, completion_tokens=46.4]
validation: 47%|████▋ | 189/400 [00:17<00:02, 93.58it/s, reward=-1.97, num_turns=1.34, num_tools=0.349, failed=0.614, completion_tokens=46.1]
validation: 48%|████▊ | 190/400 [00:17<00:02, 93.58it/s, reward=-1.94, num_turns=1.35, num_tools=0.353, failed=0.611, completion_tokens=45.8]
validation: 48%|████▊ | 191/400 [00:17<00:02, 93.58it/s, reward=-1.9, num_turns=1.35, num_tools=0.356, failed=0.607, completion_tokens=45.5]
validation: 48%|████▊ | 192/400 [00:17<00:02, 93.58it/s, reward=-1.91, num_turns=1.35, num_tools=0.359, failed=0.604, completion_tokens=45.3]
validation: 48%|████▊ | 193/400 [00:17<00:02, 93.58it/s, reward=-1.91, num_turns=1.36, num_tools=0.363, failed=0.601, completion_tokens=45.1]
validation: 48%|████▊ | 194/400 [00:17<00:02, 93.58it/s, reward=-1.91, num_turns=1.36, num_tools=0.366, failed=0.598, completion_tokens=44.9]
validation: 49%|████▉ | 195/400 [00:17<00:02, 93.58it/s, reward=-1.91, num_turns=1.36, num_tools=0.369, failed=0.595, completion_tokens=44.7]
validation: 49%|████▉ | 196/400 [00:17<00:02, 93.58it/s, reward=-1.91, num_turns=1.37, num_tools=0.372, failed=0.592, completion_tokens=44.5]
validation: 49%|████▉ | 197/400 [00:17<00:02, 77.26it/s, reward=-1.91, num_turns=1.37, num_tools=0.372, failed=0.592, completion_tokens=44.5]
validation: 49%|████▉ | 197/400 [00:17<00:02, 77.26it/s, reward=-1.91, num_turns=1.37, num_tools=0.376, failed=0.589, completion_tokens=44.4]
validation: 50%|████▉ | 198/400 [00:17<00:02, 77.26it/s, reward=-1.91, num_turns=1.37, num_tools=0.379, failed=0.586, completion_tokens=44.2]
validation: 50%|████▉ | 199/400 [00:17<00:02, 77.26it/s, reward=-1.91, num_turns=1.38, num_tools=0.382, failed=0.583, completion_tokens=44]
validation: 50%|█████ | 200/400 [00:17<00:02, 77.26it/s, reward=-1.88, num_turns=1.38, num_tools=0.385, failed=0.58, completion_tokens=43.8]
validation: 50%|█████ | 201/400 [00:17<00:02, 77.26it/s, reward=-1.85, num_turns=1.38, num_tools=0.388, failed=0.577, completion_tokens=43.6]
validation: 50%|█████ | 202/400 [00:17<00:02, 77.26it/s, reward=-1.82, num_turns=1.39, num_tools=0.391, failed=0.574, completion_tokens=43.4]
validation: 51%|█████ | 203/400 [00:17<00:02, 77.26it/s, reward=-1.79, num_turns=1.39, num_tools=0.394, failed=0.571, completion_tokens=43.3]
validation: 51%|█████ | 204/400 [00:17<00:02, 77.26it/s, reward=-1.8, num_turns=1.39, num_tools=0.397, failed=0.574, completion_tokens=43.2]
validation: 51%|█████▏ | 205/400 [00:17<00:02, 77.26it/s, reward=-1.8, num_turns=1.4, num_tools=0.4, failed=0.571, completion_tokens=43.2]
validation: 52%|█████▏ | 206/400 [00:17<00:02, 77.26it/s, reward=-1.81, num_turns=1.4, num_tools=0.403, failed=0.573, completion_tokens=43.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 52%|█████▏ | 207/400 [00:17<00:02, 77.26it/s, reward=-1.81, num_turns=1.4, num_tools=0.406, failed=0.57, completion_tokens=42.9]
validation: 52%|█████▏ | 208/400 [00:17<00:02, 77.26it/s, reward=-1.81, num_turns=1.4, num_tools=0.404, failed=0.572, completion_tokens=42.9]
validation: 52%|█████▏ | 209/400 [00:17<00:02, 77.26it/s, reward=-1.82, num_turns=1.4, num_tools=0.402, failed=0.569, completion_tokens=42.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 52%|█████▎ | 210/400 [00:17<00:02, 82.28it/s, reward=-1.82, num_turns=1.4, num_tools=0.402, failed=0.569, completion_tokens=42.9]
validation: 52%|█████▎ | 210/400 [00:17<00:02, 82.28it/s, reward=-1.81, num_turns=1.4, num_tools=0.405, failed=0.567, completion_tokens=42.8]
validation: 53%|█████▎ | 211/400 [00:17<00:02, 82.28it/s, reward=-1.81, num_turns=1.4, num_tools=0.403, failed=0.569, completion_tokens=42.8]
validation: 53%|█████▎ | 212/400 [00:17<00:02, 82.28it/s, reward=-1.82, num_turns=1.4, num_tools=0.401, failed=0.571, completion_tokens=42.8]
validation: 53%|█████▎ | 213/400 [00:17<00:02, 82.28it/s, reward=-1.79, num_turns=1.4, num_tools=0.404, failed=0.568, completion_tokens=42.6]
validation: 54%|█████▎ | 214/400 [00:17<00:02, 82.28it/s, reward=-1.76, num_turns=1.4, num_tools=0.407, failed=0.565, completion_tokens=42.5]
validation: 54%|█████▍ | 215/400 [00:17<00:02, 82.28it/s, reward=-1.74, num_turns=1.4, num_tools=0.409, failed=0.563, completion_tokens=42.3]
validation: 54%|█████▍ | 216/400 [00:17<00:02, 82.28it/s, reward=-1.74, num_turns=1.41, num_tools=0.412, failed=0.56, completion_tokens=42.3]
validation: 54%|█████▍ | 217/400 [00:17<00:02, 82.28it/s, reward=-1.74, num_turns=1.41, num_tools=0.415, failed=0.558, completion_tokens=42.1]
validation: 55%|█████▍ | 218/400 [00:17<00:02, 82.28it/s, reward=-1.71, num_turns=1.41, num_tools=0.417, failed=0.555, completion_tokens=42]
validation: 55%|█████▍ | 219/400 [00:17<00:02, 82.28it/s, reward=-1.69, num_turns=1.42, num_tools=0.42, failed=0.553, completion_tokens=41.9]
validation: 55%|█████▌ | 220/400 [00:17<00:02, 82.28it/s, reward=-1.67, num_turns=1.42, num_tools=0.423, failed=0.55, completion_tokens=41.9]
validation: 55%|█████▌ | 221/400 [00:17<00:02, 82.28it/s, reward=-1.67, num_turns=1.42, num_tools=0.425, failed=0.548, completion_tokens=41.8]
validation: 56%|█████▌ | 222/400 [00:17<00:02, 82.28it/s, reward=-1.68, num_turns=1.42, num_tools=0.428, failed=0.545, completion_tokens=41.7]
validation: 56%|█████▌ | 223/400 [00:17<00:02, 80.59it/s, reward=-1.68, num_turns=1.42, num_tools=0.428, failed=0.545, completion_tokens=41.7]
validation: 56%|█████▌ | 223/400 [00:17<00:02, 80.59it/s, reward=-1.68, num_turns=1.43, num_tools=0.43, failed=0.543, completion_tokens=41.6]
validation: 56%|█████▌ | 224/400 [00:17<00:02, 80.59it/s, reward=-1.68, num_turns=1.43, num_tools=0.433, failed=0.54, completion_tokens=41.4]
validation: 56%|█████▋ | 225/400 [00:17<00:02, 80.59it/s, reward=-1.66, num_turns=1.43, num_tools=0.436, failed=0.538, completion_tokens=41.3]
validation: 56%|█████▋ | 226/400 [00:17<00:02, 80.59it/s, reward=-1.65, num_turns=1.43, num_tools=0.438, failed=0.535, completion_tokens=41.1]
validation: 57%|█████▋ | 227/400 [00:17<00:02, 80.59it/s, reward=-1.65, num_turns=1.44, num_tools=0.441, failed=0.533, completion_tokens=40.9]
validation: 57%|█████▋ | 228/400 [00:17<00:02, 80.59it/s, reward=-1.65, num_turns=1.44, num_tools=0.443, failed=0.531, completion_tokens=40.8]
validation: 57%|█████▋ | 229/400 [00:17<00:02, 80.59it/s, reward=-1.65, num_turns=1.44, num_tools=0.445, failed=0.528, completion_tokens=40.7]
validation: 57%|█████▊ | 230/400 [00:17<00:02, 80.59it/s, reward=-1.65, num_turns=1.44, num_tools=0.448, failed=0.526, completion_tokens=40.6]
validation: 58%|█████▊ | 231/400 [00:18<00:02, 80.59it/s, reward=-1.65, num_turns=1.45, num_tools=0.45, failed=0.524, completion_tokens=40.5]
validation: 58%|█████▊ | 232/400 [00:18<00:02, 80.59it/s, reward=-1.66, num_turns=1.45, num_tools=0.453, failed=0.522, completion_tokens=40.3]
validation: 58%|█████▊ | 233/400 [00:18<00:02, 80.59it/s, reward=-1.63, num_turns=1.45, num_tools=0.455, failed=0.519, completion_tokens=40.2]
validation: 58%|█████▊ | 234/400 [00:18<00:02, 78.61it/s, reward=-1.63, num_turns=1.45, num_tools=0.455, failed=0.519, completion_tokens=40.2]
validation: 58%|█████▊ | 234/400 [00:18<00:02, 78.61it/s, reward=-1.62, num_turns=1.45, num_tools=0.457, failed=0.517, completion_tokens=40]
validation: 59%|█████▉ | 235/400 [00:18<00:02, 78.61it/s, reward=-1.6, num_turns=1.46, num_tools=0.46, failed=0.515, completion_tokens=39.9]
validation: 59%|█████▉ | 236/400 [00:18<00:02, 78.61it/s, reward=-1.59, num_turns=1.46, num_tools=0.462, failed=0.513, completion_tokens=39.7]
validation: 59%|█████▉ | 237/400 [00:18<00:02, 78.61it/s, reward=-1.57, num_turns=1.46, num_tools=0.464, failed=0.511, completion_tokens=39.6]
validation: 60%|█████▉ | 238/400 [00:18<00:02, 78.61it/s, reward=-1.56, num_turns=1.46, num_tools=0.466, failed=0.508, completion_tokens=39.4]
validation: 60%|█████▉ | 239/400 [00:18<00:02, 78.61it/s, reward=-1.56, num_turns=1.46, num_tools=0.469, failed=0.506, completion_tokens=39.4]
validation: 60%|██████ | 240/400 [00:18<00:02, 78.61it/s, reward=-1.53, num_turns=1.47, num_tools=0.471, failed=0.504, completion_tokens=39.3]
validation: 60%|██████ | 241/400 [00:18<00:02, 78.61it/s, reward=-1.52, num_turns=1.47, num_tools=0.473, failed=0.502, completion_tokens=39.2]
validation: 60%|██████ | 242/400 [00:18<00:02, 78.61it/s, reward=-1.5, num_turns=1.47, num_tools=0.475, failed=0.5, completion_tokens=39.1]
validation: 61%|██████ | 243/400 [00:18<00:01, 78.61it/s, reward=-1.5, num_turns=1.47, num_tools=0.477, failed=0.498, completion_tokens=39.1]
validation: 61%|██████ | 244/400 [00:18<00:01, 78.61it/s, reward=-1.48, num_turns=1.48, num_tools=0.48, failed=0.496, completion_tokens=38.9]
validation: 61%|██████▏ | 245/400 [00:18<00:01, 78.61it/s, reward=-1.48, num_turns=1.48, num_tools=0.482, failed=0.494, completion_tokens=38.8]
validation: 62%|██████▏ | 246/400 [00:18<00:01, 78.61it/s, reward=-1.46, num_turns=1.48, num_tools=0.484, failed=0.492, completion_tokens=38.7]
validation: 62%|██████▏ | 247/400 [00:18<00:01, 78.61it/s, reward=-1.45, num_turns=1.48, num_tools=0.486, failed=0.49, completion_tokens=38.7]
validation: 62%|██████▏ | 248/400 [00:18<00:01, 78.61it/s, reward=-1.45, num_turns=1.48, num_tools=0.488, failed=0.488, completion_tokens=38.5]
validation: 62%|██████▏ | 249/400 [00:18<00:01, 88.18it/s, reward=-1.45, num_turns=1.48, num_tools=0.488, failed=0.488, completion_tokens=38.5]
validation: 62%|██████▏ | 249/400 [00:18<00:01, 88.18it/s, reward=-1.46, num_turns=1.49, num_tools=0.49, failed=0.486, completion_tokens=38.3]
validation: 62%|██████▎ | 250/400 [00:18<00:01, 88.18it/s, reward=-1.44, num_turns=1.49, num_tools=0.492, failed=0.484, completion_tokens=38.4]
validation: 63%|██████▎ | 251/400 [00:18<00:01, 88.18it/s, reward=-1.44, num_turns=1.49, num_tools=0.494, failed=0.482, completion_tokens=38.4]
validation: 63%|██████▎ | 252/400 [00:18<00:01, 88.18it/s, reward=-1.45, num_turns=1.49, num_tools=0.496, failed=0.48, completion_tokens=38.3]
validation: 63%|██████▎ | 253/400 [00:18<00:01, 88.18it/s, reward=-1.45, num_turns=1.49, num_tools=0.498, failed=0.478, completion_tokens=38.3]
validation: 64%|██████▎ | 254/400 [00:18<00:01, 88.18it/s, reward=-1.45, num_turns=1.5, num_tools=0.5, failed=0.476, completion_tokens=38.2]
validation: 64%|██████▍ | 255/400 [00:18<00:01, 88.18it/s, reward=-1.44, num_turns=1.5, num_tools=0.502, failed=0.475, completion_tokens=38.1]
validation: 64%|██████▍ | 256/400 [00:18<00:01, 88.18it/s, reward=-1.44, num_turns=1.5, num_tools=0.504, failed=0.473, completion_tokens=38]
validation: 64%|██████▍ | 257/400 [00:18<00:01, 88.18it/s, reward=-1.43, num_turns=1.5, num_tools=0.506, failed=0.471, completion_tokens=37.9]
validation: 64%|██████▍ | 258/400 [00:18<00:01, 88.18it/s, reward=-1.41, num_turns=1.5, num_tools=0.508, failed=0.469, completion_tokens=37.8]
validation: 65%|██████▍ | 259/400 [00:18<00:01, 88.18it/s, reward=-1.4, num_turns=1.51, num_tools=0.51, failed=0.467, completion_tokens=37.8]
validation: 65%|██████▌ | 260/400 [00:18<00:01, 88.18it/s, reward=-1.4, num_turns=1.51, num_tools=0.512, failed=0.465, completion_tokens=37.7]
validation: 65%|██████▌ | 261/400 [00:18<00:01, 88.18it/s, reward=-1.39, num_turns=1.51, num_tools=0.513, failed=0.464, completion_tokens=37.6]
validation: 66%|██████▌ | 262/400 [00:18<00:01, 88.18it/s, reward=-1.39, num_turns=1.51, num_tools=0.515, failed=0.462, completion_tokens=37.6]
validation: 66%|██████▌ | 263/400 [00:18<00:01, 88.18it/s, reward=-1.38, num_turns=1.51, num_tools=0.517, failed=0.46, completion_tokens=37.5]
validation: 66%|██████▌ | 264/400 [00:18<00:01, 88.18it/s, reward=-1.38, num_turns=1.52, num_tools=0.519, failed=0.458, completion_tokens=37.4]
validation: 66%|██████▋ | 265/400 [00:18<00:01, 88.18it/s, reward=-1.36, num_turns=1.52, num_tools=0.521, failed=0.457, completion_tokens=37.4]
validation: 66%|██████▋ | 266/400 [00:18<00:01, 88.18it/s, reward=-1.36, num_turns=1.52, num_tools=0.523, failed=0.455, completion_tokens=37.3]
validation: 67%|██████▋ | 267/400 [00:18<00:01, 88.18it/s, reward=-1.34, num_turns=1.52, num_tools=0.524, failed=0.453, completion_tokens=37.2]
validation: 67%|██████▋ | 268/400 [00:18<00:01, 88.18it/s, reward=-1.32, num_turns=1.52, num_tools=0.526, failed=0.451, completion_tokens=37.1]
validation: 67%|██████▋ | 269/400 [00:18<00:01, 88.18it/s, reward=-1.32, num_turns=1.52, num_tools=0.528, failed=0.45, completion_tokens=37.1]
validation: 68%|██████▊ | 270/400 [00:18<00:01, 88.18it/s, reward=-1.31, num_turns=1.53, num_tools=0.53, failed=0.448, completion_tokens=37]
validation: 68%|██████▊ | 271/400 [00:18<00:01, 88.18it/s, reward=-1.3, num_turns=1.53, num_tools=0.531, failed=0.446, completion_tokens=37][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 68%|██████▊ | 272/400 [00:18<00:01, 88.18it/s, reward=-1.31, num_turns=1.53, num_tools=0.533, failed=0.449, completion_tokens=37]
validation: 68%|██████▊ | 273/400 [00:18<00:01, 114.79it/s, reward=-1.31, num_turns=1.53, num_tools=0.533, failed=0.449, completion_tokens=37]
validation: 68%|██████▊ | 273/400 [00:18<00:01, 114.79it/s, reward=-1.29, num_turns=1.53, num_tools=0.535, failed=0.447, completion_tokens=37]
validation: 68%|██████▊ | 274/400 [00:18<00:01, 114.79it/s, reward=-1.3, num_turns=1.53, num_tools=0.536, failed=0.445, completion_tokens=37]
validation: 69%|██████▉ | 275/400 [00:18<00:01, 114.79it/s, reward=-1.3, num_turns=1.53, num_tools=0.538, failed=0.444, completion_tokens=36.9]
validation: 69%|██████▉ | 276/400 [00:18<00:01, 114.79it/s, reward=-1.3, num_turns=1.54, num_tools=0.54, failed=0.442, completion_tokens=36.8]
validation: 69%|██████▉ | 277/400 [00:18<00:01, 114.79it/s, reward=-1.31, num_turns=1.54, num_tools=0.542, failed=0.44, completion_tokens=37]
validation: 70%|██████▉ | 278/400 [00:18<00:01, 114.79it/s, reward=-1.31, num_turns=1.54, num_tools=0.543, failed=0.439, completion_tokens=36.9]
validation: 70%|██████▉ | 279/400 [00:18<00:01, 114.79it/s, reward=-1.31, num_turns=1.54, num_tools=0.545, failed=0.437, completion_tokens=36.8]
validation: 70%|███████ | 280/400 [00:18<00:01, 114.79it/s, reward=-1.3, num_turns=1.54, num_tools=0.546, failed=0.436, completion_tokens=36.8]
validation: 70%|███████ | 281/400 [00:18<00:01, 114.79it/s, reward=-1.3, num_turns=1.54, num_tools=0.548, failed=0.434, completion_tokens=36.7]
validation: 70%|███████ | 282/400 [00:18<00:01, 114.79it/s, reward=-1.31, num_turns=1.55, num_tools=0.55, failed=0.433, completion_tokens=36.7]
validation: 71%|███████ | 283/400 [00:18<00:01, 114.79it/s, reward=-1.31, num_turns=1.55, num_tools=0.551, failed=0.431, completion_tokens=36.7]
validation: 71%|███████ | 284/400 [00:18<00:01, 114.79it/s, reward=-1.3, num_turns=1.55, num_tools=0.553, failed=0.43, completion_tokens=36.7]
validation: 71%|███████▏ | 285/400 [00:18<00:01, 114.79it/s, reward=-1.29, num_turns=1.55, num_tools=0.554, failed=0.428, completion_tokens=36.6]
validation: 72%|███████▏ | 286/400 [00:18<00:00, 114.79it/s, reward=-1.3, num_turns=1.55, num_tools=0.556, failed=0.427, completion_tokens=36.5]
validation: 72%|███████▏ | 287/400 [00:18<00:00, 114.79it/s, reward=-1.28, num_turns=1.55, num_tools=0.557, failed=0.425, completion_tokens=36.4]
validation: 72%|███████▏ | 288/400 [00:18<00:00, 114.79it/s, reward=-1.29, num_turns=1.56, num_tools=0.559, failed=0.424, completion_tokens=36.5]
validation: 72%|███████▏ | 289/400 [00:18<00:00, 114.79it/s, reward=-1.29, num_turns=1.56, num_tools=0.561, failed=0.422, completion_tokens=36.5]
validation: 72%|███████▎ | 290/400 [00:18<00:00, 114.79it/s, reward=-1.29, num_turns=1.56, num_tools=0.562, failed=0.421, completion_tokens=36.5]
validation: 73%|███████▎ | 291/400 [00:18<00:00, 114.79it/s, reward=-1.29, num_turns=1.56, num_tools=0.564, failed=0.419, completion_tokens=36.4]
validation: 73%|███████▎ | 292/400 [00:18<00:00, 114.79it/s, reward=-1.3, num_turns=1.56, num_tools=0.565, failed=0.418, completion_tokens=36.5]
validation: 73%|███████▎ | 293/400 [00:18<00:00, 114.79it/s, reward=-1.29, num_turns=1.56, num_tools=0.567, failed=0.416, completion_tokens=36.8]
validation: 74%|███████▎ | 294/400 [00:18<00:00, 114.79it/s, reward=-1.28, num_turns=1.56, num_tools=0.568, failed=0.415, completion_tokens=36.7]
validation: 74%|███████▍ | 295/400 [00:18<00:00, 136.76it/s, reward=-1.28, num_turns=1.56, num_tools=0.568, failed=0.415, completion_tokens=36.7]
validation: 74%|███████▍ | 295/400 [00:18<00:00, 136.76it/s, reward=-1.27, num_turns=1.57, num_tools=0.569, failed=0.414, completion_tokens=36.7]
validation: 74%|███████▍ | 296/400 [00:18<00:00, 136.76it/s, reward=-1.26, num_turns=1.57, num_tools=0.571, failed=0.412, completion_tokens=36.6]
validation: 74%|███████▍ | 297/400 [00:18<00:00, 136.76it/s, reward=-1.26, num_turns=1.57, num_tools=0.572, failed=0.411, completion_tokens=36.5]
validation: 74%|███████▍ | 298/400 [00:18<00:00, 136.76it/s, reward=-1.24, num_turns=1.57, num_tools=0.574, failed=0.409, completion_tokens=36.5]
validation: 75%|███████▍ | 299/400 [00:18<00:00, 136.76it/s, reward=-1.23, num_turns=1.57, num_tools=0.575, failed=0.408, completion_tokens=36.4]
validation: 75%|███████▌ | 300/400 [00:18<00:00, 136.76it/s, reward=-1.23, num_turns=1.57, num_tools=0.577, failed=0.407, completion_tokens=36.4]
validation: 75%|███████▌ | 301/400 [00:18<00:00, 136.76it/s, reward=-1.23, num_turns=1.57, num_tools=0.578, failed=0.405, completion_tokens=36.3]
validation: 76%|███████▌ | 302/400 [00:18<00:00, 136.76it/s, reward=-1.22, num_turns=1.58, num_tools=0.579, failed=0.404, completion_tokens=36.2]
validation: 76%|███████▌ | 303/400 [00:18<00:00, 136.76it/s, reward=-1.21, num_turns=1.58, num_tools=0.581, failed=0.403, completion_tokens=36.2]
validation: 76%|███████▌ | 304/400 [00:18<00:00, 136.76it/s, reward=-1.21, num_turns=1.58, num_tools=0.582, failed=0.401, completion_tokens=36.1]
validation: 76%|███████▋ | 305/400 [00:18<00:00, 136.76it/s, reward=-1.2, num_turns=1.58, num_tools=0.584, failed=0.4, completion_tokens=36]
validation: 76%|███████▋ | 306/400 [00:18<00:00, 136.76it/s, reward=-1.19, num_turns=1.58, num_tools=0.585, failed=0.399, completion_tokens=36]
validation: 77%|███████▋ | 307/400 [00:18<00:00, 136.76it/s, reward=-1.18, num_turns=1.58, num_tools=0.586, failed=0.397, completion_tokens=35.9]
validation: 77%|███████▋ | 308/400 [00:18<00:00, 136.76it/s, reward=-1.18, num_turns=1.58, num_tools=0.588, failed=0.396, completion_tokens=35.8]
validation: 77%|███████▋ | 309/400 [00:18<00:00, 136.76it/s, reward=-1.18, num_turns=1.59, num_tools=0.589, failed=0.395, completion_tokens=35.9]
validation: 78%|███████▊ | 310/400 [00:18<00:00, 136.76it/s, reward=-1.18, num_turns=1.59, num_tools=0.59, failed=0.394, completion_tokens=35.8]
validation: 78%|███████▊ | 311/400 [00:18<00:00, 136.76it/s, reward=-1.19, num_turns=1.59, num_tools=0.592, failed=0.392, completion_tokens=35.8]
validation: 78%|███████▊ | 312/400 [00:18<00:00, 136.76it/s, reward=-1.17, num_turns=1.59, num_tools=0.593, failed=0.391, completion_tokens=35.8]
validation: 78%|███████▊ | 313/400 [00:18<00:00, 136.76it/s, reward=-1.18, num_turns=1.59, num_tools=0.594, failed=0.39, completion_tokens=35.8]
validation: 78%|███████▊ | 314/400 [00:18<00:00, 136.76it/s, reward=-1.16, num_turns=1.59, num_tools=0.596, failed=0.389, completion_tokens=35.7]
validation: 79%|███████▉ | 315/400 [00:18<00:00, 136.76it/s, reward=-1.16, num_turns=1.59, num_tools=0.597, failed=0.387, completion_tokens=35.7]
validation: 79%|███████▉ | 316/400 [00:18<00:00, 136.76it/s, reward=-1.15, num_turns=1.59, num_tools=0.598, failed=0.386, completion_tokens=35.6]
validation: 79%|███████▉ | 317/400 [00:18<00:00, 136.76it/s, reward=-1.14, num_turns=1.6, num_tools=0.599, failed=0.385, completion_tokens=35.6]
validation: 80%|███████▉ | 318/400 [00:18<00:00, 152.02it/s, reward=-1.14, num_turns=1.6, num_tools=0.599, failed=0.385, completion_tokens=35.6]
validation: 80%|███████▉ | 318/400 [00:18<00:00, 152.02it/s, reward=-1.12, num_turns=1.6, num_tools=0.601, failed=0.384, completion_tokens=35.5]
validation: 80%|███████▉ | 319/400 [00:18<00:00, 152.02it/s, reward=-1.11, num_turns=1.6, num_tools=0.602, failed=0.382, completion_tokens=35.5]
validation: 80%|████████ | 320/400 [00:18<00:00, 152.02it/s, reward=-1.09, num_turns=1.6, num_tools=0.603, failed=0.381, completion_tokens=35.5]
validation: 80%|████████ | 321/400 [00:18<00:00, 152.02it/s, reward=-1.07, num_turns=1.6, num_tools=0.604, failed=0.38, completion_tokens=35.4]
validation: 80%|████████ | 322/400 [00:18<00:00, 152.02it/s, reward=-1.07, num_turns=1.6, num_tools=0.606, failed=0.379, completion_tokens=35.5]
validation: 81%|████████ | 323/400 [00:18<00:00, 152.02it/s, reward=-1.05, num_turns=1.6, num_tools=0.607, failed=0.378, completion_tokens=35.4]
validation: 81%|████████ | 324/400 [00:18<00:00, 152.02it/s, reward=-1.06, num_turns=1.6, num_tools=0.608, failed=0.377, completion_tokens=35.5]
validation: 81%|████████▏ | 325/400 [00:18<00:00, 152.02it/s, reward=-1.05, num_turns=1.61, num_tools=0.609, failed=0.375, completion_tokens=35.5]
validation: 82%|████████▏ | 326/400 [00:18<00:00, 152.02it/s, reward=-1.06, num_turns=1.61, num_tools=0.61, failed=0.374, completion_tokens=35.4]
validation: 82%|████████▏ | 327/400 [00:18<00:00, 152.02it/s, reward=-1.06, num_turns=1.61, num_tools=0.612, failed=0.373, completion_tokens=35.4]
validation: 82%|████████▏ | 328/400 [00:18<00:00, 152.02it/s, reward=-1.06, num_turns=1.61, num_tools=0.613, failed=0.372, completion_tokens=35.4]
validation: 82%|████████▏ | 329/400 [00:18<00:00, 152.02it/s, reward=-1.06, num_turns=1.61, num_tools=0.614, failed=0.371, completion_tokens=35.4]
validation: 82%|████████▎ | 330/400 [00:18<00:00, 152.02it/s, reward=-1.07, num_turns=1.61, num_tools=0.615, failed=0.37, completion_tokens=35.4]
validation: 83%|████████▎ | 331/400 [00:18<00:00, 152.02it/s, reward=-1.07, num_turns=1.61, num_tools=0.616, failed=0.369, completion_tokens=35.5]
validation: 83%|████████▎ | 332/400 [00:18<00:00, 152.02it/s, reward=-1.07, num_turns=1.61, num_tools=0.617, failed=0.367, completion_tokens=35.5]
validation: 83%|████████▎ | 333/400 [00:18<00:00, 152.02it/s, reward=-1.07, num_turns=1.62, num_tools=0.619, failed=0.366, completion_tokens=35.5]
validation: 84%|████████▎ | 334/400 [00:18<00:00, 152.02it/s, reward=-1.07, num_turns=1.62, num_tools=0.62, failed=0.365, completion_tokens=35.5]
validation: 84%|████████▍ | 335/400 [00:18<00:00, 152.02it/s, reward=-1.08, num_turns=1.61, num_tools=0.618, failed=0.364, completion_tokens=35.8]
validation: 84%|████████▍ | 336/400 [00:18<00:00, 147.35it/s, reward=-1.08, num_turns=1.61, num_tools=0.618, failed=0.364, completion_tokens=35.8]
validation: 84%|████████▍ | 336/400 [00:18<00:00, 147.35it/s, reward=-1.08, num_turns=1.62, num_tools=0.619, failed=0.363, completion_tokens=35.8]
validation: 84%|████████▍ | 337/400 [00:18<00:00, 147.35it/s, reward=-1.07, num_turns=1.62, num_tools=0.62, failed=0.362, completion_tokens=36]
validation: 84%|████████▍ | 338/400 [00:18<00:00, 147.35it/s, reward=-1.07, num_turns=1.62, num_tools=0.621, failed=0.361, completion_tokens=36]
validation: 85%|████████▍ | 339/400 [00:18<00:00, 147.35it/s, reward=-1.07, num_turns=1.62, num_tools=0.622, failed=0.36, completion_tokens=36.1]
validation: 85%|████████▌ | 340/400 [00:18<00:00, 147.35it/s, reward=-1.07, num_turns=1.62, num_tools=0.624, failed=0.359, completion_tokens=36.1]
validation: 85%|████████▌ | 341/400 [00:18<00:00, 147.35it/s, reward=-1.07, num_turns=1.62, num_tools=0.625, failed=0.358, completion_tokens=36.1]
validation: 86%|████████▌ | 342/400 [00:18<00:00, 147.35it/s, reward=-1.06, num_turns=1.62, num_tools=0.626, failed=0.357, completion_tokens=36.2]
validation: 86%|████████▌ | 343/400 [00:18<00:00, 147.35it/s, reward=-1.06, num_turns=1.62, num_tools=0.627, failed=0.356, completion_tokens=36.2]
validation: 86%|████████▌ | 344/400 [00:18<00:00, 147.35it/s, reward=-1.06, num_turns=1.62, num_tools=0.628, failed=0.355, completion_tokens=36.3]
validation: 86%|████████▋ | 345/400 [00:18<00:00, 147.35it/s, reward=-1.06, num_turns=1.63, num_tools=0.629, failed=0.354, completion_tokens=36.3]
validation: 86%|████████▋ | 346/400 [00:18<00:00, 147.35it/s, reward=-1.06, num_turns=1.63, num_tools=0.63, failed=0.353, completion_tokens=36.4]
validation: 87%|████████▋ | 347/400 [00:18<00:00, 147.35it/s, reward=-1.05, num_turns=1.63, num_tools=0.634, failed=0.352, completion_tokens=36.5]
validation: 87%|████████▋ | 348/400 [00:18<00:00, 147.35it/s, reward=-1.04, num_turns=1.63, num_tools=0.635, failed=0.351, completion_tokens=36.5]
validation: 87%|████████▋ | 349/400 [00:18<00:00, 147.35it/s, reward=-1.04, num_turns=1.63, num_tools=0.636, failed=0.35, completion_tokens=36.5]
validation: 88%|████████▊ | 350/400 [00:18<00:00, 147.35it/s, reward=-1.04, num_turns=1.63, num_tools=0.637, failed=0.349, completion_tokens=36.5]
validation: 88%|████████▊ | 351/400 [00:18<00:00, 147.35it/s, reward=-1.03, num_turns=1.63, num_tools=0.638, failed=0.348, completion_tokens=36.5]
validation: 88%|████████▊ | 352/400 [00:18<00:00, 147.35it/s, reward=-1.04, num_turns=1.63, num_tools=0.639, failed=0.347, completion_tokens=36.6]
validation: 88%|████████▊ | 353/400 [00:18<00:00, 151.15it/s, reward=-1.04, num_turns=1.63, num_tools=0.639, failed=0.347, completion_tokens=36.6]
validation: 88%|████████▊ | 353/400 [00:18<00:00, 151.15it/s, reward=-1.04, num_turns=1.63, num_tools=0.64, failed=0.346, completion_tokens=36.6]
validation: 88%|████████▊ | 354/400 [00:18<00:00, 151.15it/s, reward=-1.04, num_turns=1.64, num_tools=0.641, failed=0.345, completion_tokens=36.6]
validation: 89%|████████▉ | 355/400 [00:18<00:00, 151.15it/s, reward=-1.05, num_turns=1.64, num_tools=0.642, failed=0.344, completion_tokens=36.6]
validation: 89%|████████▉ | 356/400 [00:18<00:00, 151.15it/s, reward=-1.05, num_turns=1.64, num_tools=0.643, failed=0.343, completion_tokens=36.6]
validation: 89%|████████▉ | 357/400 [00:18<00:00, 151.15it/s, reward=-1.05, num_turns=1.64, num_tools=0.644, failed=0.342, completion_tokens=36.7]
validation: 90%|████████▉ | 358/400 [00:18<00:00, 151.15it/s, reward=-1.05, num_turns=1.64, num_tools=0.645, failed=0.341, completion_tokens=36.7]
validation: 90%|████████▉ | 359/400 [00:18<00:00, 151.15it/s, reward=-1.06, num_turns=1.64, num_tools=0.649, failed=0.34, completion_tokens=36.6]
validation: 90%|█████████ | 360/400 [00:18<00:00, 151.15it/s, reward=-1.06, num_turns=1.64, num_tools=0.647, failed=0.339, completion_tokens=37]
validation: 90%|█████████ | 361/400 [00:18<00:00, 151.15it/s, reward=-1.06, num_turns=1.64, num_tools=0.648, failed=0.338, completion_tokens=37.1]
validation: 90%|█████████ | 362/400 [00:18<00:00, 151.15it/s, reward=-1.07, num_turns=1.64, num_tools=0.649, failed=0.337, completion_tokens=37.2]
validation: 91%|█████████ | 363/400 [00:18<00:00, 151.15it/s, reward=-1.07, num_turns=1.64, num_tools=0.65, failed=0.336, completion_tokens=37.3]
validation: 91%|█████████ | 364/400 [00:18<00:00, 151.15it/s, reward=-1.07, num_turns=1.65, num_tools=0.651, failed=0.335, completion_tokens=37.3]
validation: 91%|█████████▏| 365/400 [00:18<00:00, 151.15it/s, reward=-1.07, num_turns=1.65, num_tools=0.652, failed=0.334, completion_tokens=37.5]
validation: 92%|█████████▏| 366/400 [00:18<00:00, 151.15it/s, reward=-1.08, num_turns=1.65, num_tools=0.653, failed=0.333, completion_tokens=37.6]
validation: 92%|█████████▏| 367/400 [00:18<00:00, 151.15it/s, reward=-1.07, num_turns=1.65, num_tools=0.654, failed=0.332, completion_tokens=37.8]
validation: 92%|█████████▏| 368/400 [00:18<00:00, 151.15it/s, reward=-1.07, num_turns=1.65, num_tools=0.655, failed=0.332, completion_tokens=37.9]
validation: 92%|█████████▏| 369/400 [00:18<00:00, 151.15it/s, reward=-1.07, num_turns=1.65, num_tools=0.656, failed=0.331, completion_tokens=37.9]
validation: 92%|█████████▎| 370/400 [00:18<00:00, 120.71it/s, reward=-1.07, num_turns=1.65, num_tools=0.656, failed=0.331, completion_tokens=37.9]
validation: 92%|█████████▎| 370/400 [00:18<00:00, 120.71it/s, reward=-1.07, num_turns=1.65, num_tools=0.657, failed=0.33, completion_tokens=38.2]
validation: 93%|█████████▎| 371/400 [00:18<00:00, 120.71it/s, reward=-1.07, num_turns=1.65, num_tools=0.658, failed=0.329, completion_tokens=38.2]
validation: 93%|█████████▎| 372/400 [00:18<00:00, 120.71it/s, reward=-1.07, num_turns=1.65, num_tools=0.659, failed=0.328, completion_tokens=38.4]
validation: 93%|█████████▎| 373/400 [00:18<00:00, 120.71it/s, reward=-1.08, num_turns=1.65, num_tools=0.66, failed=0.327, completion_tokens=38.5]
validation: 94%|█████████▎| 374/400 [00:18<00:00, 120.71it/s, reward=-1.08, num_turns=1.66, num_tools=0.66, failed=0.326, completion_tokens=38.5]
validation: 94%|█████████▍| 375/400 [00:18<00:00, 120.71it/s, reward=-1.08, num_turns=1.66, num_tools=0.661, failed=0.325, completion_tokens=38.7]
validation: 94%|█████████▍| 376/400 [00:18<00:00, 120.71it/s, reward=-1.08, num_turns=1.66, num_tools=0.662, failed=0.324, completion_tokens=38.9]
validation: 94%|█████████▍| 377/400 [00:18<00:00, 120.71it/s, reward=-1.08, num_turns=1.66, num_tools=0.663, failed=0.324, completion_tokens=38.9]
validation: 94%|█████████▍| 378/400 [00:19<00:00, 120.71it/s, reward=-1.08, num_turns=1.66, num_tools=0.664, failed=0.323, completion_tokens=39.1]
validation: 95%|█████████▍| 379/400 [00:19<00:00, 120.71it/s, reward=-1.08, num_turns=1.66, num_tools=0.665, failed=0.322, completion_tokens=39.1]
validation: 95%|█████████▌| 380/400 [00:19<00:00, 120.71it/s, reward=-1.09, num_turns=1.66, num_tools=0.666, failed=0.321, completion_tokens=39.4]
validation: 95%|█████████▌| 381/400 [00:19<00:00, 120.71it/s, reward=-1.09, num_turns=1.66, num_tools=0.667, failed=0.32, completion_tokens=39.5]
validation: 96%|█████████▌| 382/400 [00:19<00:00, 120.71it/s, reward=-1.09, num_turns=1.66, num_tools=0.668, failed=0.319, completion_tokens=39.6]
validation: 96%|█████████▌| 383/400 [00:19<00:00, 120.71it/s, reward=-1.09, num_turns=1.66, num_tools=0.668, failed=0.319, completion_tokens=39.8]
validation: 96%|█████████▌| 384/400 [00:19<00:00, 90.59it/s, reward=-1.09, num_turns=1.66, num_tools=0.668, failed=0.319, completion_tokens=39.8]
validation: 96%|█████████▌| 384/400 [00:19<00:00, 90.59it/s, reward=-1.1, num_turns=1.66, num_tools=0.669, failed=0.318, completion_tokens=40]
validation: 96%|█████████▋| 385/400 [00:19<00:00, 90.59it/s, reward=-1.1, num_turns=1.67, num_tools=0.673, failed=0.317, completion_tokens=40.1]
validation: 96%|█████████▋| 386/400 [00:19<00:00, 90.59it/s, reward=-1.09, num_turns=1.67, num_tools=0.674, failed=0.316, completion_tokens=40.3]
validation: 97%|█████████▋| 387/400 [00:19<00:00, 90.59it/s, reward=-1.09, num_turns=1.67, num_tools=0.674, failed=0.315, completion_tokens=40.6]
validation: 97%|█████████▋| 388/400 [00:19<00:00, 90.59it/s, reward=-1.1, num_turns=1.67, num_tools=0.675, failed=0.314, completion_tokens=40.8]
validation: 97%|█████████▋| 389/400 [00:19<00:00, 90.59it/s, reward=-1.1, num_turns=1.67, num_tools=0.679, failed=0.314, completion_tokens=41.4]
validation: 98%|█████████▊| 390/400 [00:20<00:00, 90.59it/s, reward=-1.1, num_turns=1.67, num_tools=0.682, failed=0.313, completion_tokens=41.6]
validation: 98%|█████████▊| 391/400 [00:20<00:00, 90.59it/s, reward=-1.1, num_turns=1.67, num_tools=0.683, failed=0.312, completion_tokens=42]
validation: 98%|█████████▊| 392/400 [00:20<00:00, 90.59it/s, reward=-1.1, num_turns=1.67, num_tools=0.684, failed=0.311, completion_tokens=42.4]
validation: 98%|█████████▊| 393/400 [00:20<00:00, 90.59it/s, reward=-1.1, num_turns=1.68, num_tools=0.69, failed=0.31, completion_tokens=42.6]
validation: 98%|█████████▊| 394/400 [00:20<00:00, 90.59it/s, reward=-1.11, num_turns=1.68, num_tools=0.688, failed=0.31, completion_tokens=44.2]
validation: 99%|█████████▉| 395/400 [00:20<00:00, 90.59it/s, reward=-1.11, num_turns=1.68, num_tools=0.689, failed=0.309, completion_tokens=44.8]
validation: 99%|█████████▉| 396/400 [00:20<00:00, 24.89it/s, reward=-1.11, num_turns=1.68, num_tools=0.689, failed=0.309, completion_tokens=44.8]
validation: 99%|█████████▉| 396/400 [00:20<00:00, 24.89it/s, reward=-1.11, num_turns=1.68, num_tools=0.687, failed=0.308, completion_tokens=46.4]
validation: 99%|█████████▉| 397/400 [00:20<00:00, 24.89it/s, reward=-1.12, num_turns=1.68, num_tools=0.688, failed=0.307, completion_tokens=47.1]
validation: 100%|█████████▉| 398/400 [00:20<00:00, 24.89it/s, reward=-1.12, num_turns=1.68, num_tools=0.686, failed=0.307, completion_tokens=48.7]
validation: 100%|█████████▉| 399/400 [00:20<00:00, 24.89it/s, reward=-1.12, num_turns=1.68, num_tools=0.689, failed=0.306, completion_tokens=49.1]
validation: 100%|██████████| 400/400 [00:20<00:00, 24.89it/s, reward=-1.12, num_turns=1.68, num_tools=0.69, failed=0.305, completion_tokens=49.8]
validation: 100%|██████████| 400/400 [00:20<00:00, 19.19it/s, reward=-1.12, num_turns=1.68, num_tools=0.69, failed=0.305, completion_tokens=49.8]
Val avg reward: -1.124
============================================================
Step 11/50
============================================================
step 11: 0%| | 0/32 [00:00<?, ?it/s]
step 11: 3%|▎ | 1/32 [00:00<00:30, 1.02it/s]
step 11: 3%|▎ | 1/32 [00:00<00:30, 1.02it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=5]
step 11: 6%|▋ | 2/32 [00:01<00:18, 1.59it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=5]
step 11: 6%|▋ | 2/32 [00:01<00:18, 1.59it/s, reward=-2.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=14.5]
step 11: 9%|▉ | 3/32 [00:01<00:18, 1.59it/s, reward=-2.33, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=18]
step 11: 12%|█▎ | 4/32 [00:01<00:17, 1.59it/s, reward=-2.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=28.8]
step 11: 16%|█▌ | 5/32 [00:01<00:16, 1.59it/s, reward=-2.4, num_turns=1.6, num_tools=0.6, failed=0, completion_tokens=28.2]
step 11: 19%|█▉ | 6/32 [00:01<00:16, 1.59it/s, reward=-1.83, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=27.2]
step 11: 22%|██▏ | 7/32 [00:01<00:15, 1.59it/s, reward=-1.43, num_turns=1.71, num_tools=0.714, failed=0, completion_tokens=27]
step 11: 25%|██▌ | 8/32 [00:01<00:15, 1.59it/s, reward=-1.5, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=26.4]
step 11: 28%|██▊ | 9/32 [00:01<00:14, 1.59it/s, reward=-1.56, num_turns=1.78, num_tools=0.778, failed=0, completion_tokens=26]
step 11: 31%|███▏ | 10/32 [00:01<00:13, 1.59it/s, reward=-1.6, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=26.1]
step 11: 34%|███▍ | 11/32 [00:01<00:13, 1.59it/s, reward=-1.64, num_turns=1.82, num_tools=0.818, failed=0, completion_tokens=26.6]
step 11: 38%|███▊ | 12/32 [00:01<00:12, 1.59it/s, reward=-1.67, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=26.7]
step 11: 41%|████ | 13/32 [00:01<00:01, 14.27it/s, reward=-1.67, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=26.7]
step 11: 41%|████ | 13/32 [00:01<00:01, 14.27it/s, reward=-1.69, num_turns=1.85, num_tools=0.846, failed=0, completion_tokens=26.9]
step 11: 44%|████▍ | 14/32 [00:01<00:01, 14.27it/s, reward=-1.71, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=27.2]
step 11: 47%|████▋ | 15/32 [00:01<00:01, 14.27it/s, reward=-1.73, num_turns=1.87, num_tools=0.867, failed=0, completion_tokens=27.5]
step 11: 50%|█████ | 16/32 [00:01<00:01, 14.27it/s, reward=-1.54, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=28]
step 11: 53%|█████▎ | 17/32 [00:01<00:01, 14.27it/s, reward=-1.57, num_turns=1.88, num_tools=0.882, failed=0, completion_tokens=28.6]
step 11: 56%|█████▋ | 18/32 [00:01<00:00, 14.27it/s, reward=-1.59, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=29.1]
step 11: 59%|█████▉ | 19/32 [00:01<00:00, 14.27it/s, reward=-1.46, num_turns=1.89, num_tools=0.895, failed=0, completion_tokens=29.6]
step 11: 62%|██████▎ | 20/32 [00:01<00:00, 14.27it/s, reward=-1.33, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=30.1]
step 11: 66%|██████▌ | 21/32 [00:01<00:00, 14.27it/s, reward=-1.37, num_turns=1.9, num_tools=0.905, failed=0, completion_tokens=30.3]
step 11: 69%|██████▉ | 22/32 [00:01<00:00, 14.27it/s, reward=-1.39, num_turns=1.91, num_tools=0.909, failed=0, completion_tokens=30.5]
step 11: 72%|███████▏ | 23/32 [00:01<00:00, 14.27it/s, reward=-1.42, num_turns=1.91, num_tools=0.913, failed=0, completion_tokens=30.7]
step 11: 75%|███████▌ | 24/32 [00:01<00:00, 14.27it/s, reward=-1.44, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=30.9]
step 11: 78%|███████▊ | 25/32 [00:01<00:00, 29.24it/s, reward=-1.44, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=30.9]
step 11: 78%|███████▊ | 25/32 [00:01<00:00, 29.24it/s, reward=-1.33, num_turns=1.92, num_tools=0.92, failed=0, completion_tokens=31.3]
step 11: 81%|████████▏ | 26/32 [00:01<00:00, 29.24it/s, reward=-1.24, num_turns=1.92, num_tools=0.923, failed=0, completion_tokens=31.9]
step 11: 84%|████████▍ | 27/32 [00:01<00:00, 29.24it/s, reward=-1.17, num_turns=1.93, num_tools=0.926, failed=0, completion_tokens=32.5]
step 11: 88%|████████▊ | 28/32 [00:01<00:00, 29.24it/s, reward=-1.23, num_turns=1.89, num_tools=0.893, failed=0, completion_tokens=35.7]
step 11: 91%|█████████ | 29/32 [00:01<00:00, 29.24it/s, reward=-1.26, num_turns=1.9, num_tools=0.897, failed=0, completion_tokens=36.5]
step 11: 94%|█████████▍| 30/32 [00:01<00:00, 29.24it/s, reward=-1.28, num_turns=1.9, num_tools=0.933, failed=0, completion_tokens=37.2]
step 11: 97%|█████████▋| 31/32 [00:02<00:00, 29.24it/s, reward=-1.31, num_turns=1.9, num_tools=0.935, failed=0, completion_tokens=39.4]
step 11: 100%|██████████| 32/32 [00:02<00:00, 29.24it/s, reward=-1.36, num_turns=1.88, num_tools=0.906, failed=0, completion_tokens=48.7]
step 11: 100%|██████████| 32/32 [00:02<00:00, 12.45it/s, reward=-1.36, num_turns=1.88, num_tools=0.906, failed=0, completion_tokens=48.7]
group 0: mean=-1.50 std=1.509 min=-3.0 max=+1.3 | What is France's population density in people per
group 1: mean=-0.94 std=1.667 min=-3.0 max=+1.5 | What is the distance from Earth to the Sun in km i
group 2: mean=-2.12 std=0.331 min=-3.0 max=-2.0 | What is the GDP of France?
group 3: mean=-0.88 std=1.452 min=-2.0 max=+1.0 | How old was Guido van Rossum in 2020?
Avg reward: -1.359 | Avg tools/rollout: 0.9 | groups with variance: 4/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0009
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0005
Packed 32 trajectories into 3 sequences of length 2048
train: 0%| | 0/3 [00:00<?, ?it/s]
train: 33%|███▎ | 1/3 [00:01<00:03, 1.99s/it]
train: 33%|███▎ | 1/3 [00:01<00:03, 1.99s/it, loss/train=-0.711, loss/grad_norm=1.94, loss/learning_rate=5e-5, loss/entropy=0.804]
train: 67%|██████▋ | 2/3 [00:02<00:01, 1.02s/it, loss/train=-0.711, loss/grad_norm=1.94, loss/learning_rate=5e-5, loss/entropy=0.804]
train: 67%|██████▋ | 2/3 [00:02<00:01, 1.02s/it, loss/train=0.395, loss/grad_norm=2.08, loss/learning_rate=5e-5, loss/entropy=1.18]
train: 100%|██████████| 3/3 [00:02<00:00, 1.41it/s, loss/train=0.395, loss/grad_norm=2.08, loss/learning_rate=5e-5, loss/entropy=1.18]
train: 100%|██████████| 3/3 [00:02<00:00, 1.41it/s, loss/train=-0.169, loss/grad_norm=1.78, loss/learning_rate=5e-5, loss/entropy=1.04](APIServer pid=14938) Adapters before cleanup: ['default']
(APIServer pid=14938) Keeping active adapter(s): ['default']
(APIServer pid=14938) Adapters after cleanup: ['default']
train: 100%|██████████| 3/3 [00:30<00:00, 10.23s/it, loss/train=-0.169, loss/grad_norm=1.78, loss/learning_rate=5e-5, loss/entropy=1.04]
============================================================
Step 12/50
============================================================
step 12: 0%| | 0/32 [00:00<?, ?it/s]
step 12: 3%|▎ | 1/32 [00:01<00:38, 1.25s/it]
step 12: 3%|▎ | 1/32 [00:01<00:38, 1.25s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=40]
step 12: 6%|▋ | 2/32 [00:01<00:17, 1.68it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=40]
step 12: 6%|▋ | 2/32 [00:01<00:17, 1.68it/s, reward=0.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=31.2]
step 12: 9%|▉ | 3/32 [00:01<00:17, 1.68it/s, reward=-0.333, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=28.5]
step 12: 12%|█▎ | 4/32 [00:01<00:16, 1.68it/s, reward=0.75, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=26.6]
step 12: 16%|█▌ | 5/32 [00:01<00:16, 1.68it/s, reward=1.4, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=25.7]
step 12: 19%|█▉ | 6/32 [00:01<00:15, 1.68it/s, reward=1.83, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=25.1]
step 12: 22%|██▏ | 7/32 [00:01<00:14, 1.68it/s, reward=2.14, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=24.7]
step 12: 25%|██▌ | 8/32 [00:01<00:14, 1.68it/s, reward=2, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=24.8]
step 12: 28%|██▊ | 9/32 [00:01<00:13, 1.68it/s, reward=1.89, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=24.6]
step 12: 31%|███▏ | 10/32 [00:01<00:13, 1.68it/s, reward=1.5, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=24.7]
step 12: 34%|███▍ | 11/32 [00:01<00:12, 1.68it/s, reward=1.41, num_turns=1.91, num_tools=0.909, failed=0, completion_tokens=24.7]
step 12: 38%|███▊ | 12/32 [00:01<00:11, 1.68it/s, reward=1.33, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=24.8]
step 12: 41%|████ | 13/32 [00:01<00:11, 1.68it/s, reward=1.27, num_turns=1.92, num_tools=0.923, failed=0, completion_tokens=25]
step 12: 44%|████▍ | 14/32 [00:01<00:10, 1.68it/s, reward=1.21, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=25.1]
step 12: 47%|████▋ | 15/32 [00:01<00:10, 1.68it/s, reward=1.17, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=25.3]
step 12: 50%|█████ | 16/32 [00:01<00:09, 1.68it/s, reward=1.12, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=25.5]
step 12: 53%|█████▎ | 17/32 [00:01<00:08, 1.68it/s, reward=0.941, num_turns=1.94, num_tools=0.941, failed=0, completion_tokens=25.7]
step 12: 56%|█████▋ | 18/32 [00:01<00:00, 21.05it/s, reward=0.941, num_turns=1.94, num_tools=0.941, failed=0, completion_tokens=25.7]
step 12: 56%|█████▋ | 18/32 [00:01<00:00, 21.05it/s, reward=0.778, num_turns=1.94, num_tools=0.944, failed=0, completion_tokens=26.1]
step 12: 59%|█████▉ | 19/32 [00:01<00:00, 21.05it/s, reward=0.842, num_turns=1.95, num_tools=0.947, failed=0, completion_tokens=26.6]
step 12: 62%|██████▎ | 20/32 [00:01<00:00, 21.05it/s, reward=1, num_turns=1.95, num_tools=0.95, failed=0, completion_tokens=27.1]
step 12: 66%|██████▌ | 21/32 [00:01<00:00, 21.05it/s, reward=1.14, num_turns=1.95, num_tools=0.952, failed=0, completion_tokens=26.9]
step 12: 69%|██████▉ | 22/32 [00:01<00:00, 21.05it/s, reward=1, num_turns=1.95, num_tools=0.955, failed=0, completion_tokens=27.6]
step 12: 72%|███████▏ | 23/32 [00:01<00:00, 21.05it/s, reward=0.87, num_turns=1.96, num_tools=0.957, failed=0, completion_tokens=28.3]
step 12: 75%|███████▌ | 24/32 [00:01<00:00, 21.05it/s, reward=1, num_turns=1.96, num_tools=0.958, failed=0, completion_tokens=28.1]
step 12: 78%|███████▊ | 25/32 [00:01<00:00, 21.05it/s, reward=0.96, num_turns=1.96, num_tools=0.96, failed=0, completion_tokens=29]
step 12: 81%|████████▏ | 26/32 [00:01<00:00, 24.55it/s, reward=0.96, num_turns=1.96, num_tools=0.96, failed=0, completion_tokens=29]
step 12: 81%|████████▏ | 26/32 [00:01<00:00, 24.55it/s, reward=0.962, num_turns=1.96, num_tools=0.962, failed=0, completion_tokens=28.9]
step 12: 84%|████████▍ | 27/32 [00:01<00:00, 24.55it/s, reward=0.852, num_turns=1.96, num_tools=0.963, failed=0, completion_tokens=29.9]
step 12: 88%|████████▊ | 28/32 [00:01<00:00, 24.55it/s, reward=0.75, num_turns=1.96, num_tools=0.964, failed=0, completion_tokens=30.8]
step 12: 91%|█████████ | 29/32 [00:01<00:00, 24.55it/s, reward=0.655, num_turns=1.97, num_tools=0.966, failed=0, completion_tokens=31.8]
step 12: 94%|█████████▍| 30/32 [00:01<00:00, 24.55it/s, reward=0.567, num_turns=1.97, num_tools=0.967, failed=0, completion_tokens=32.2]
step 12: 97%|█████████▋| 31/32 [00:02<00:00, 24.55it/s, reward=0.452, num_turns=1.94, num_tools=0.935, failed=0, completion_tokens=40]
step 12: 100%|██████████| 32/32 [00:03<00:00, 10.05it/s, reward=0.452, num_turns=1.94, num_tools=0.935, failed=0, completion_tokens=40]
step 12: 100%|██████████| 32/32 [00:03<00:00, 10.05it/s, reward=0.484, num_turns=1.97, num_tools=0.969, failed=0, completion_tokens=43.4]
step 12: 100%|██████████| 32/32 [00:03<00:00, 10.12it/s, reward=0.484, num_turns=1.97, num_tools=0.969, failed=0, completion_tokens=43.4]
group 0: mean=-0.44 std=1.861 min=-3.0 max=+2.0 | What is the distance from Earth to the Sun in km i
group 1: mean=+4.00 std=0.000 min=+4.0 max=+4.0 | What's the weather like in Sydney?
group 2: mean=+0.50 std=0.250 min=+0.0 max=+1.0 | What is the temperature in Berlin in Fahrenheit?
group 3: mean=-2.12 std=0.331 min=-3.0 max=-2.0 | What is the speed of light?
Avg reward: 0.484 | Avg tools/rollout: 1.0 | groups with variance: 3/4
Packed 18 trajectories into 2 sequences of length 2048
train: 0%| | 0/2 [00:00<?, ?it/s]
train: 50%|█████ | 1/2 [00:01<00:01, 1.95s/it]
train: 50%|█████ | 1/2 [00:01<00:01, 1.95s/it, loss/train=-0.202, loss/grad_norm=1.38, loss/learning_rate=5e-5, loss/entropy=0.754]
train: 100%|██████████| 2/2 [00:02<00:00, 1.00s/it, loss/train=-0.202, loss/grad_norm=1.38, loss/learning_rate=5e-5, loss/entropy=0.754]
train: 100%|██████████| 2/2 [00:02<00:00, 1.00s/it, loss/train=-1.73, loss/grad_norm=1.7, loss/learning_rate=5e-5, loss/entropy=1.34] (APIServer pid=14938) Adapters before cleanup: ['default']
(APIServer pid=14938) Keeping active adapter(s): ['default']
(APIServer pid=14938) Adapters after cleanup: ['default']
train: 100%|██████████| 2/2 [00:30<00:00, 15.35s/it, loss/train=-1.73, loss/grad_norm=1.7, loss/learning_rate=5e-5, loss/entropy=1.34]
============================================================
Step 13/50
============================================================
step 13: 0%| | 0/32 [00:00<?, ?it/s]
step 13: 3%|▎ | 1/32 [00:01<00:37, 1.22s/it]
step 13: 3%|▎ | 1/32 [00:01<00:37, 1.22s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=36]
step 13: 6%|▋ | 2/32 [00:01<00:36, 1.22s/it, reward=-2.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=28]
step 13: 9%|▉ | 3/32 [00:01<00:10, 2.79it/s, reward=-2.5, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=28]
step 13: 9%|▉ | 3/32 [00:01<00:10, 2.79it/s, reward=-0.333, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=26.2]
step 13: 12%|█▎ | 4/32 [00:01<00:10, 2.79it/s, reward=0.125, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=25.2]
step 13: 16%|█▌ | 5/32 [00:01<00:09, 2.79it/s, reward=-0.3, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=24.8]
step 13: 19%|█▉ | 6/32 [00:01<00:09, 2.79it/s, reward=0.0833, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=24.4]
step 13: 22%|██▏ | 7/32 [00:01<00:08, 2.79it/s, reward=0.143, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=24.5]
step 13: 25%|██▌ | 8/32 [00:01<00:08, 2.79it/s, reward=-0.125, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=24.6]
step 13: 28%|██▊ | 9/32 [00:01<00:08, 2.79it/s, reward=0.111, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=24.3]
step 13: 31%|███▏ | 10/32 [00:01<00:07, 2.79it/s, reward=0.15, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=24.5]
step 13: 34%|███▍ | 11/32 [00:01<00:07, 2.79it/s, reward=-0.0455, num_turns=1.91, num_tools=0.909, failed=0, completion_tokens=24.7]
step 13: 38%|███▊ | 12/32 [00:01<00:07, 2.79it/s, reward=0, num_turns=1.92, num_tools=0.917, failed=0, completion_tokens=24.8]
step 13: 41%|████ | 13/32 [00:01<00:06, 2.79it/s, reward=0.0385, num_turns=1.92, num_tools=0.923, failed=0, completion_tokens=25]
step 13: 44%|████▍ | 14/32 [00:01<00:06, 2.79it/s, reward=-0.107, num_turns=1.93, num_tools=0.929, failed=0, completion_tokens=25.2]
step 13: 47%|████▋ | 15/32 [00:01<00:06, 2.79it/s, reward=-0.0667, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=25.4]
step 13: 50%|█████ | 16/32 [00:01<00:00, 19.08it/s, reward=-0.0667, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=25.4]
step 13: 50%|█████ | 16/32 [00:01<00:00, 19.08it/s, reward=-0.188, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=25.8]
step 13: 53%|█████▎ | 17/32 [00:01<00:00, 19.08it/s, reward=-0.147, num_turns=1.94, num_tools=0.941, failed=0, completion_tokens=26.1]
step 13: 56%|█████▋ | 18/32 [00:01<00:00, 19.08it/s, reward=-0.25, num_turns=1.94, num_tools=0.944, failed=0, completion_tokens=26.6]
step 13: 59%|█████▉ | 19/32 [00:01<00:00, 19.08it/s, reward=-0.132, num_turns=1.95, num_tools=0.947, failed=0, completion_tokens=26.4]
step 13: 62%|██████▎ | 20/32 [00:01<00:00, 19.08it/s, reward=-0.025, num_turns=1.95, num_tools=0.95, failed=0, completion_tokens=26.2]
step 13: 66%|██████▌ | 21/32 [00:01<00:00, 19.08it/s, reward=0.0714, num_turns=1.95, num_tools=0.952, failed=0, completion_tokens=26]
step 13: 69%|██████▉ | 22/32 [00:01<00:00, 19.08it/s, reward=0.0985, num_turns=1.95, num_tools=0.955, failed=0, completion_tokens=26.8]
step 13: 72%|███████▏ | 23/32 [00:01<00:00, 26.42it/s, reward=0.0985, num_turns=1.95, num_tools=0.955, failed=0, completion_tokens=26.8]
step 13: 72%|███████▏ | 23/32 [00:01<00:00, 26.42it/s, reward=0.00725, num_turns=1.96, num_tools=0.957, failed=0, completion_tokens=27.5]
step 13: 75%|███████▌ | 24/32 [00:01<00:00, 26.42it/s, reward=-0.0764, num_turns=1.96, num_tools=0.958, failed=0, completion_tokens=28]
step 13: 78%|███████▊ | 25/32 [00:01<00:00, 26.42it/s, reward=0.00667, num_turns=1.96, num_tools=0.96, failed=0, completion_tokens=27.7]
step 13: 81%|████████▏ | 26/32 [00:01<00:00, 26.42it/s, reward=0.0641, num_turns=1.96, num_tools=0.962, failed=0, completion_tokens=27.8]
step 13: 84%|████████▍ | 27/32 [00:01<00:00, 26.42it/s, reward=-0.0123, num_turns=1.96, num_tools=0.963, failed=0, completion_tokens=27.8]
step 13: 88%|████████▊ | 28/32 [00:01<00:00, 26.42it/s, reward=0.0238, num_turns=1.96, num_tools=0.964, failed=0, completion_tokens=28.6]
step 13: 91%|█████████ | 29/32 [00:01<00:00, 26.42it/s, reward=-0.0805, num_turns=1.93, num_tools=0.931, failed=0, completion_tokens=32.3]
step 13: 94%|█████████▍| 30/32 [00:02<00:00, 11.53it/s, reward=-0.0805, num_turns=1.93, num_tools=0.931, failed=0, completion_tokens=32.3]
step 13: 94%|█████████▍| 30/32 [00:02<00:00, 11.53it/s, reward=-0.0333, num_turns=1.93, num_tools=0.933, failed=0, completion_tokens=37.5]
step 13: 97%|█████████▋| 31/32 [00:03<00:00, 11.53it/s, reward=0.0215, num_turns=1.97, num_tools=0.968, failed=0, completion_tokens=40.6]
step 13: 100%|██████████| 32/32 [00:03<00:00, 11.53it/s, reward=-0.0729, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=54.7]
step 13: 100%|██████████| 32/32 [00:03<00:00, 9.95it/s, reward=-0.0729, num_turns=1.94, num_tools=0.938, failed=0, completion_tokens=54.7]
group 0: mean=-2.12 std=0.331 min=-3.0 max=-2.0 | Which is hotter right now, Tokyo or Cairo?
group 1: mean=+0.75 std=0.433 min=+0.5 max=+1.5 | What is the temperature in London in Fahrenheit?
group 2: mean=-0.67 std=1.886 min=-3.0 max=+1.7 | What is Germany's population density in people per
group 3: mean=+1.75 std=1.561 min=-2.0 max=+4.0 | What is 575 times 22?
Avg reward: -0.073 | Avg tools/rollout: 0.9 | groups with variance: 4/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0011
Packed 32 trajectories into 3 sequences of length 2048
train: 0%| | 0/3 [00:00<?, ?it/s]
train: 33%|███▎ | 1/3 [00:01<00:03, 1.94s/it]
train: 33%|███▎ | 1/3 [00:01<00:03, 1.94s/it, loss/train=0.274, loss/grad_norm=1.74, loss/learning_rate=5e-5, loss/entropy=0.954]
train: 67%|██████▋ | 2/3 [00:02<00:00, 1.00it/s, loss/train=0.274, loss/grad_norm=1.74, loss/learning_rate=5e-5, loss/entropy=0.954]
train: 67%|██████▋ | 2/3 [00:02<00:00, 1.00it/s, loss/train=0.162, loss/grad_norm=1.79, loss/learning_rate=5e-5, loss/entropy=0.522]
train: 100%|██████████| 3/3 [00:02<00:00, 1.44it/s, loss/train=0.162, loss/grad_norm=1.79, loss/learning_rate=5e-5, loss/entropy=0.522]
train: 100%|██████████| 3/3 [00:02<00:00, 1.44it/s, loss/train=-1.37, loss/grad_norm=0.978, loss/learning_rate=5e-5, loss/entropy=0.731](APIServer pid=14938) Adapters before cleanup: ['default']
(APIServer pid=14938) Keeping active adapter(s): ['default']
(APIServer pid=14938) Adapters after cleanup: ['default']
train: 100%|██████████| 3/3 [00:30<00:00, 10.25s/it, loss/train=-1.37, loss/grad_norm=0.978, loss/learning_rate=5e-5, loss/entropy=0.731]
============================================================
Step 14/50
============================================================
step 14: 0%| | 0/32 [00:00<?, ?it/s]
step 14: 3%|▎ | 1/32 [00:01<00:43, 1.39s/it]
step 14: 3%|▎ | 1/32 [00:01<00:43, 1.39s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=21]
step 14: 6%|▋ | 2/32 [00:01<00:41, 1.39s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=21]
step 14: 9%|▉ | 3/32 [00:01<00:40, 1.39s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=21.3]
step 14: 12%|█▎ | 4/32 [00:01<00:38, 1.39s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=21.6]
step 14: 16%|█▌ | 5/32 [00:01<00:37, 1.39s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=21.9]
step 14: 19%|█▉ | 6/32 [00:01<00:36, 1.39s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=22.3]
step 14: 22%|██▏ | 7/32 [00:01<00:34, 1.39s/it, reward=4, num_turns=2, num_tools=1, failed=0, completion_tokens=22.7]
step 14: 25%|██▌ | 8/32 [00:01<00:33, 1.39s/it, reward=3.75, num_turns=2, num_tools=1, failed=0, completion_tokens=23.2]
step 14: 28%|██▊ | 9/32 [00:01<00:31, 1.39s/it, reward=3.11, num_turns=2, num_tools=1, failed=0, completion_tokens=23.2]
step 14: 31%|███▏ | 10/32 [00:01<00:30, 1.39s/it, reward=3.13, num_turns=2, num_tools=1, failed=0, completion_tokens=23.7]
step 14: 34%|███▍ | 11/32 [00:01<00:29, 1.39s/it, reward=3.15, num_turns=2, num_tools=1, failed=0, completion_tokens=24.1]
step 14: 38%|███▊ | 12/32 [00:01<00:27, 1.39s/it, reward=3.17, num_turns=2, num_tools=1, failed=0, completion_tokens=24.5]
step 14: 41%|████ | 13/32 [00:01<00:26, 1.39s/it, reward=3, num_turns=2, num_tools=1, failed=0, completion_tokens=24.8]
step 14: 44%|████▍ | 14/32 [00:01<00:01, 12.73it/s, reward=3, num_turns=2, num_tools=1, failed=0, completion_tokens=24.8]
step 14: 44%|████▍ | 14/32 [00:01<00:01, 12.73it/s, reward=3.07, num_turns=2, num_tools=1, failed=0, completion_tokens=25.1]
step 14: 47%|████▋ | 15/32 [00:01<00:01, 12.73it/s, reward=3.13, num_turns=2, num_tools=1, failed=0, completion_tokens=25.3]
step 14: 50%|█████ | 16/32 [00:01<00:01, 12.73it/s, reward=3.19, num_turns=2, num_tools=1, failed=0, completion_tokens=25.5]
step 14: 53%|█████▎ | 17/32 [00:01<00:01, 12.73it/s, reward=3.24, num_turns=2, num_tools=1, failed=0, completion_tokens=25.7]
step 14: 56%|█████▋ | 18/32 [00:01<00:01, 12.73it/s, reward=3.28, num_turns=2, num_tools=1, failed=0, completion_tokens=25.9]
step 14: 59%|█████▉ | 19/32 [00:01<00:01, 12.73it/s, reward=3.13, num_turns=2, num_tools=1, failed=0, completion_tokens=25.9]
step 14: 62%|██████▎ | 20/32 [00:01<00:00, 12.73it/s, reward=3, num_turns=2, num_tools=1, failed=0, completion_tokens=26]
step 14: 66%|██████▌ | 21/32 [00:01<00:00, 12.73it/s, reward=2.88, num_turns=2, num_tools=1, failed=0, completion_tokens=26]
step 14: 69%|██████▉ | 22/32 [00:01<00:00, 12.73it/s, reward=2.77, num_turns=2, num_tools=1, failed=0, completion_tokens=26.1]
step 14: 72%|███████▏ | 23/32 [00:01<00:00, 12.73it/s, reward=2.72, num_turns=2, num_tools=1, failed=0, completion_tokens=26]
step 14: 75%|███████▌ | 24/32 [00:01<00:00, 12.73it/s, reward=2.62, num_turns=2, num_tools=1, failed=0, completion_tokens=26.2]
step 14: 78%|███████▊ | 25/32 [00:01<00:00, 12.73it/s, reward=2.68, num_turns=2, num_tools=1, failed=0, completion_tokens=26.1]
step 14: 81%|████████▏ | 26/32 [00:01<00:00, 12.73it/s, reward=2.64, num_turns=2, num_tools=1, failed=0, completion_tokens=26.4]
step 14: 84%|████████▍ | 27/32 [00:01<00:00, 25.95it/s, reward=2.64, num_turns=2, num_tools=1, failed=0, completion_tokens=26.4]
step 14: 84%|████████▍ | 27/32 [00:01<00:00, 25.95it/s, reward=2.43, num_turns=1.96, num_tools=0.963, failed=0, completion_tokens=29.6]
step 14: 88%|████████▊ | 28/32 [00:01<00:00, 25.95it/s, reward=2.27, num_turns=1.96, num_tools=0.964, failed=0, completion_tokens=30]
step 14: 91%|█████████ | 29/32 [00:01<00:00, 25.95it/s, reward=2.13, num_turns=1.97, num_tools=0.966, failed=0, completion_tokens=30.2]
step 14: 94%|█████████▍| 30/32 [00:01<00:00, 25.95it/s, reward=1.99, num_turns=1.97, num_tools=0.967, failed=0, completion_tokens=30.8]
step 14: 97%|█████████▋| 31/32 [00:01<00:00, 25.95it/s, reward=1.86, num_turns=1.97, num_tools=0.968, failed=0, completion_tokens=32]
step 14: 100%|██████████| 32/32 [00:01<00:00, 25.95it/s, reward=1.74, num_turns=1.97, num_tools=0.969, failed=0, completion_tokens=33.1]
step 14: 100%|██████████| 32/32 [00:01<00:00, 17.14it/s, reward=1.74, num_turns=1.97, num_tools=0.969, failed=0, completion_tokens=33.1]
group 0: mean=+4.00 std=0.000 min=+4.0 max=+4.0 | What's the weather like in Paris?
group 1: mean=+3.75 std=0.323 min=+3.3 max=+4.0 | Convert 28 kg to lbs.
group 2: mean=-1.67 std=1.302 min=-3.0 max=+1.7 | What is the population of Germany divided by its a
group 3: mean=+0.88 std=0.545 min=+0.5 max=+2.0 | What is the temperature in Tokyo in Fahrenheit?
Avg reward: 1.740 | Avg tools/rollout: 1.0 | groups with variance: 3/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0012
Packed 24 trajectories into 2 sequences of length 2048
train: 0%| | 0/2 [00:00<?, ?it/s]
train: 50%|█████ | 1/2 [00:01<00:01, 1.83s/it]
train: 50%|█████ | 1/2 [00:01<00:01, 1.83s/it, loss/train=-0.223, loss/grad_norm=0.761, loss/learning_rate=5e-5, loss/entropy=0.571]
train: 100%|██████████| 2/2 [00:02<00:00, 1.07it/s, loss/train=-0.223, loss/grad_norm=0.761, loss/learning_rate=5e-5, loss/entropy=0.571]
train: 100%|██████████| 2/2 [00:02<00:00, 1.07it/s, loss/train=0.383, loss/grad_norm=2.75, loss/learning_rate=5e-5, loss/entropy=0.603] (APIServer pid=14938) Adapters before cleanup: ['default']
(APIServer pid=14938) Keeping active adapter(s): ['default']
(APIServer pid=14938) Adapters after cleanup: ['default']
train: 100%|██████████| 2/2 [00:29<00:00, 14.67s/it, loss/train=0.383, loss/grad_norm=2.75, loss/learning_rate=5e-5, loss/entropy=0.603]
============================================================
Step 15/50
============================================================
step 15: 0%| | 0/32 [00:00<?, ?it/s]
step 15: 3%|▎ | 1/32 [00:01<00:42, 1.38s/it]
step 15: 3%|▎ | 1/32 [00:01<00:42, 1.38s/it, reward=1, num_turns=2, num_tools=1, failed=0, completion_tokens=22.5]
step 15: 6%|▋ | 2/32 [00:01<00:41, 1.38s/it, reward=1.5, num_turns=2, num_tools=1, failed=0, completion_tokens=22.2]
step 15: 9%|▉ | 3/32 [00:01<00:40, 1.38s/it, reward=1.67, num_turns=2, num_tools=1, failed=0, completion_tokens=22.2]
step 15: 12%|█▎ | 4/32 [00:01<00:38, 1.38s/it, reward=1.75, num_turns=2, num_tools=1, failed=0, completion_tokens=22.1]
step 15: 16%|█▌ | 5/32 [00:01<00:37, 1.38s/it, reward=1.8, num_turns=2, num_tools=1, failed=0, completion_tokens=22.1]
step 15: 19%|█▉ | 6/32 [00:01<00:35, 1.38s/it, reward=1.17, num_turns=2, num_tools=1, failed=0, completion_tokens=22.9]
step 15: 22%|██▏ | 7/32 [00:01<00:34, 1.38s/it, reward=0.714, num_turns=2, num_tools=1, failed=0, completion_tokens=23.5]
step 15: 25%|██▌ | 8/32 [00:01<00:33, 1.38s/it, reward=1.12, num_turns=2, num_tools=1, failed=0, completion_tokens=24.1]
step 15: 28%|██▊ | 9/32 [00:01<00:31, 1.38s/it, reward=1.44, num_turns=2, num_tools=1, failed=0, completion_tokens=24.6]
step 15: 31%|███▏ | 10/32 [00:01<00:30, 1.38s/it, reward=1.1, num_turns=2, num_tools=1, failed=0, completion_tokens=24.9]
step 15: 34%|███▍ | 11/32 [00:01<00:29, 1.38s/it, reward=0.818, num_turns=2, num_tools=1, failed=0, completion_tokens=25]
step 15: 38%|███▊ | 12/32 [00:01<00:27, 1.38s/it, reward=1.08, num_turns=2, num_tools=1, failed=0, completion_tokens=25.2]
step 15: 41%|████ | 13/32 [00:01<00:26, 1.38s/it, reward=1.31, num_turns=2, num_tools=1, failed=0, completion_tokens=25.4]
step 15: 44%|████▍ | 14/32 [00:01<00:24, 1.38s/it, reward=1.5, num_turns=2, num_tools=1, failed=0, completion_tokens=25.5]
step 15: 47%|████▋ | 15/32 [00:01<00:23, 1.38s/it, reward=1.62, num_turns=2, num_tools=1, failed=0, completion_tokens=25.7]
step 15: 50%|█████ | 16/32 [00:01<00:22, 1.38s/it, reward=1.4, num_turns=2, num_tools=1, failed=0, completion_tokens=25.9]
step 15: 53%|█████▎ | 17/32 [00:01<00:20, 1.38s/it, reward=1.55, num_turns=2, num_tools=1, failed=0, completion_tokens=26.1]
step 15: 56%|█████▋ | 18/32 [00:01<00:19, 1.38s/it, reward=1.57, num_turns=2, num_tools=1, failed=0, completion_tokens=25.9]
step 15: 59%|█████▉ | 19/32 [00:01<00:17, 1.38s/it, reward=1.6, num_turns=2, num_tools=1, failed=0, completion_tokens=25.7]
step 15: 62%|██████▎ | 20/32 [00:01<00:16, 1.38s/it, reward=1.62, num_turns=2, num_tools=1, failed=0, completion_tokens=25.5]
step 15: 66%|██████▌ | 21/32 [00:01<00:15, 1.38s/it, reward=1.63, num_turns=2, num_tools=1, failed=0, completion_tokens=25.3]
step 15: 69%|██████▉ | 22/32 [00:01<00:13, 1.38s/it, reward=1.74, num_turns=2, num_tools=1, failed=0, completion_tokens=25.4]
step 15: 72%|███████▏ | 23/32 [00:01<00:12, 1.38s/it, reward=1.58, num_turns=2, num_tools=1, failed=0, completion_tokens=25.5]
step 15: 75%|███████▌ | 24/32 [00:01<00:00, 21.63it/s, reward=1.58, num_turns=2, num_tools=1, failed=0, completion_tokens=25.5]
step 15: 75%|███████▌ | 24/32 [00:01<00:00, 21.63it/s, reward=1.43, num_turns=2, num_tools=1, failed=0, completion_tokens=26]
step 15: 78%|███████▊ | 25/32 [00:01<00:00, 21.63it/s, reward=1.29, num_turns=2, num_tools=1, failed=0, completion_tokens=26.6]
step 15: 81%|████████▏ | 26/32 [00:01<00:00, 21.63it/s, reward=1.17, num_turns=2, num_tools=1, failed=0, completion_tokens=27.2]
step 15: 84%|████████▍ | 27/32 [00:01<00:00, 21.63it/s, reward=1.05, num_turns=2, num_tools=1, failed=0, completion_tokens=27.4]
step 15: 88%|████████▊ | 28/32 [00:01<00:00, 21.63it/s, reward=1.05, num_turns=2, num_tools=1, failed=0, completion_tokens=28.8]
step 15: 91%|█████████ | 29/32 [00:01<00:00, 21.63it/s, reward=0.943, num_turns=2, num_tools=1, failed=0, completion_tokens=30.1]
step 15: 94%|█████████▍| 30/32 [00:01<00:00, 21.63it/s, reward=0.844, num_turns=2, num_tools=1, failed=0, completion_tokens=31.4]
step 15: 97%|█████████▋| 31/32 [00:01<00:00, 21.63it/s, reward=0.753, num_turns=2.03, num_tools=1.03, failed=0, completion_tokens=32.1]
step 15: 100%|██████████| 32/32 [00:02<00:00, 21.63it/s, reward=0.635, num_turns=2, num_tools=1, failed=0, completion_tokens=40.6]
step 15: 100%|██████████| 32/32 [00:02<00:00, 13.23it/s, reward=0.635, num_turns=2, num_tools=1, failed=0, completion_tokens=40.6]
group 0: mean=-2.12 std=0.331 min=-3.0 max=-2.0 | Which is hotter right now, Paris or Mumbai?
group 1: mean=-1.25 std=1.299 min=-2.0 max=+1.0 | How old was Guido van Rossum in 2020?
group 2: mean=+3.92 std=0.220 min=+3.3 max=+4.0 | Convert 26 kg to lbs.
group 3: mean=+2.00 std=0.000 min=+2.0 max=+2.0 | What is 965 plus 85?
Avg reward: 0.635 | Avg tools/rollout: 1.0 | groups with variance: 3/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0013
Packed 24 trajectories into 2 sequences of length 2048
train: 0%| | 0/2 [00:00<?, ?it/s]
train: 50%|█████ | 1/2 [00:01<00:01, 1.90s/it]
train: 50%|█████ | 1/2 [00:01<00:01, 1.90s/it, loss/train=0.154, loss/grad_norm=1.1, loss/learning_rate=5e-5, loss/entropy=0.721]
train: 100%|██████████| 2/2 [00:02<00:00, 1.02it/s, loss/train=0.154, loss/grad_norm=1.1, loss/learning_rate=5e-5, loss/entropy=0.721]
train: 100%|██████████| 2/2 [00:02<00:00, 1.02it/s, loss/train=-0.222, loss/grad_norm=6.06, loss/learning_rate=5e-5, loss/entropy=0.556](APIServer pid=14938) Adapters before cleanup: ['default']
(APIServer pid=14938) Keeping active adapter(s): ['default']
(APIServer pid=14938) Adapters after cleanup: ['default']
train: 100%|██████████| 2/2 [00:30<00:00, 15.15s/it, loss/train=-0.222, loss/grad_norm=6.06, loss/learning_rate=5e-5, loss/entropy=0.556]
Running validation...
validation: 0%| | 0/400 [00:00<?, ?it/s]
validation: 0%| | 1/400 [00:11<1:19:09, 11.90s/it]
validation: 0%| | 1/400 [00:11<1:19:09, 11.90s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=34]
validation: 0%| | 2/400 [00:11<1:18:58, 11.90s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=35]
validation: 1%| | 3/400 [00:12<21:12, 3.21s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=35]
validation: 1%| | 3/400 [00:12<21:12, 3.21s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=31]
validation: 1%| | 4/400 [00:12<21:09, 3.21s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=27.2]
validation: 1%|▏ | 5/400 [00:12<10:25, 1.58s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=27.2]
validation: 1%|▏ | 5/400 [00:12<10:25, 1.58s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=31.2]
validation: 2%|▏ | 6/400 [00:12<10:24, 1.58s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=34.2]
validation: 2%|▏ | 7/400 [00:12<10:22, 1.58s/it, reward=-2.86, num_turns=1.14, num_tools=0.143, failed=0, completion_tokens=32.4]
validation: 2%|▏ | 8/400 [00:12<10:20, 1.58s/it, reward=-2.75, num_turns=1.25, num_tools=0.25, failed=0, completion_tokens=31.7]
validation: 2%|▏ | 9/400 [00:12<10:19, 1.58s/it, reward=-2.78, num_turns=1.22, num_tools=0.222, failed=0, completion_tokens=38.1]
validation: 2%|▎ | 10/400 [00:12<03:43, 1.75it/s, reward=-2.78, num_turns=1.22, num_tools=0.222, failed=0, completion_tokens=38.1]
validation: 2%|▎ | 10/400 [00:12<03:43, 1.75it/s, reward=-2.7, num_turns=1.3, num_tools=0.3, failed=0, completion_tokens=38.3]
validation: 3%|▎ | 11/400 [00:12<03:42, 1.75it/s, reward=-2.64, num_turns=1.36, num_tools=0.364, failed=0, completion_tokens=37.4]
validation: 3%|▎ | 12/400 [00:12<03:42, 1.75it/s, reward=-2.33, num_turns=1.42, num_tools=0.417, failed=0, completion_tokens=36.2]
validation: 3%|▎ | 13/400 [00:12<03:41, 1.75it/s, reward=-2.38, num_turns=1.38, num_tools=0.385, failed=0, completion_tokens=36.5]
validation: 4%|▎ | 14/400 [00:12<03:41, 1.75it/s, reward=-1.93, num_turns=1.43, num_tools=0.429, failed=0, completion_tokens=35.9]
validation: 4%|▍ | 15/400 [00:12<03:40, 1.75it/s, reward=-1.53, num_turns=1.47, num_tools=0.467, failed=0, completion_tokens=35.5]
validation: 4%|▍ | 16/400 [00:12<03:39, 1.75it/s, reward=-1.19, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=35.1]
validation: 4%|▍ | 17/400 [00:12<03:39, 1.75it/s, reward=-0.882, num_turns=1.53, num_tools=0.529, failed=0, completion_tokens=34.7]
validation: 4%|▍ | 18/400 [00:12<03:38, 1.75it/s, reward=-0.611, num_turns=1.56, num_tools=0.556, failed=0, completion_tokens=34.4]
validation: 5%|▍ | 19/400 [00:12<03:38, 1.75it/s, reward=-0.368, num_turns=1.58, num_tools=0.579, failed=0, completion_tokens=34.1]
validation: 5%|▌ | 20/400 [00:12<03:37, 1.75it/s, reward=-0.183, num_turns=1.6, num_tools=0.6, failed=0, completion_tokens=33.9]
validation: 5%|▌ | 21/400 [00:12<03:37, 1.75it/s, reward=-0.27, num_turns=1.62, num_tools=0.619, failed=0, completion_tokens=33.8]
validation: 6%|▌ | 22/400 [00:12<03:36, 1.75it/s, reward=-0.348, num_turns=1.64, num_tools=0.636, failed=0, completion_tokens=33.7]
validation: 6%|▌ | 23/400 [00:12<03:35, 1.75it/s, reward=-0.42, num_turns=1.65, num_tools=0.652, failed=0, completion_tokens=34]
validation: 6%|▌ | 24/400 [00:12<03:35, 1.75it/s, reward=-0.486, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=34.5]
validation: 6%|▋ | 25/400 [00:12<01:01, 6.05it/s, reward=-0.486, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=34.5]
validation: 6%|▋ | 25/400 [00:12<01:01, 6.05it/s, reward=-0.587, num_turns=1.64, num_tools=0.64, failed=0, completion_tokens=34.4] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 6%|▋ | 26/400 [00:13<01:01, 6.05it/s, reward=-0.679, num_turns=1.62, num_tools=0.615, failed=0.0385, completion_tokens=34.4]
validation: 7%|▋ | 27/400 [00:13<01:01, 6.05it/s, reward=-0.765, num_turns=1.59, num_tools=0.593, failed=0.0741, completion_tokens=34.4][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 7%|▋ | 28/400 [00:13<01:01, 6.05it/s, reward=-0.845, num_turns=1.57, num_tools=0.571, failed=0.107, completion_tokens=34.4]
validation: 7%|▋ | 29/400 [00:13<00:57, 6.41it/s, reward=-0.845, num_turns=1.57, num_tools=0.571, failed=0.107, completion_tokens=34.4]
validation: 7%|▋ | 29/400 [00:13<00:57, 6.41it/s, reward=-0.92, num_turns=1.55, num_tools=0.552, failed=0.138, completion_tokens=34.4]
validation: 8%|▊ | 30/400 [00:13<00:57, 6.41it/s, reward=-0.989, num_turns=1.53, num_tools=0.533, failed=0.167, completion_tokens=34.4]
validation: 8%|▊ | 31/400 [00:13<00:57, 6.41it/s, reward=-1.02, num_turns=1.55, num_tools=0.548, failed=0.161, completion_tokens=33.6]
validation: 8%|▊ | 32/400 [00:13<00:57, 6.41it/s, reward=-1.08, num_turns=1.53, num_tools=0.531, failed=0.188, completion_tokens=33.6]
validation: 8%|▊ | 33/400 [00:13<00:57, 6.41it/s, reward=-1.11, num_turns=1.55, num_tools=0.545, failed=0.182, completion_tokens=33.2]
validation: 8%|▊ | 34/400 [00:13<00:57, 6.41it/s, reward=-1.14, num_turns=1.56, num_tools=0.559, failed=0.176, completion_tokens=32.7]
validation: 9%|▉ | 35/400 [00:13<00:56, 6.41it/s, reward=-1.08, num_turns=1.57, num_tools=0.571, failed=0.171, completion_tokens=32.4]
validation: 9%|▉ | 36/400 [00:13<00:56, 6.41it/s, reward=-0.991, num_turns=1.58, num_tools=0.583, failed=0.167, completion_tokens=32]
validation: 9%|▉ | 37/400 [00:13<00:56, 6.41it/s, reward=-0.91, num_turns=1.59, num_tools=0.595, failed=0.162, completion_tokens=31.7]
validation: 10%|▉ | 38/400 [00:13<00:56, 6.41it/s, reward=-0.833, num_turns=1.61, num_tools=0.605, failed=0.158, completion_tokens=31.4]
validation: 10%|▉ | 39/400 [00:13<00:56, 6.41it/s, reward=-0.709, num_turns=1.62, num_tools=0.615, failed=0.154, completion_tokens=31.2]
validation: 10%|█ | 40/400 [00:13<00:56, 6.41it/s, reward=-0.592, num_turns=1.62, num_tools=0.625, failed=0.15, completion_tokens=31.1]
validation: 10%|█ | 41/400 [00:13<00:56, 6.41it/s, reward=-0.48, num_turns=1.63, num_tools=0.634, failed=0.146, completion_tokens=31.1]
validation: 10%|█ | 42/400 [00:13<00:55, 6.41it/s, reward=-0.373, num_turns=1.64, num_tools=0.643, failed=0.143, completion_tokens=31]
validation: 11%|█ | 43/400 [00:13<00:55, 6.41it/s, reward=-0.411, num_turns=1.65, num_tools=0.651, failed=0.14, completion_tokens=30.8]
validation: 11%|█ | 44/400 [00:13<00:55, 6.41it/s, reward=-0.47, num_turns=1.64, num_tools=0.636, failed=0.136, completion_tokens=35.9]
validation: 11%|█▏ | 45/400 [00:13<00:55, 6.41it/s, reward=-0.504, num_turns=1.64, num_tools=0.644, failed=0.133, completion_tokens=35.6]
validation: 12%|█▏ | 46/400 [00:13<00:55, 6.41it/s, reward=-0.536, num_turns=1.65, num_tools=0.652, failed=0.13, completion_tokens=35.3]
validation: 12%|█▏ | 47/400 [00:13<00:55, 6.41it/s, reward=-0.567, num_turns=1.66, num_tools=0.66, failed=0.128, completion_tokens=35.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 12%|█▏ | 48/400 [00:13<00:23, 15.03it/s, reward=-0.567, num_turns=1.66, num_tools=0.66, failed=0.128, completion_tokens=35.1]
validation: 12%|█▏ | 48/400 [00:13<00:23, 15.03it/s, reward=-0.514, num_turns=1.67, num_tools=0.667, failed=0.125, completion_tokens=34.8]
validation: 12%|█▏ | 49/400 [00:13<00:23, 15.03it/s, reward=-0.463, num_turns=1.67, num_tools=0.673, failed=0.122, completion_tokens=34.5]
validation: 12%|█▎ | 50/400 [00:13<00:23, 15.03it/s, reward=-0.403, num_turns=1.68, num_tools=0.68, failed=0.12, completion_tokens=34.2]
validation: 13%|█▎ | 51/400 [00:13<00:23, 15.03it/s, reward=-0.356, num_turns=1.69, num_tools=0.686, failed=0.118, completion_tokens=33.9]
validation: 13%|█▎ | 52/400 [00:13<00:23, 15.03it/s, reward=-0.407, num_turns=1.67, num_tools=0.673, failed=0.135, completion_tokens=33.9]
validation: 13%|█▎ | 53/400 [00:13<00:23, 15.03it/s, reward=-0.456, num_turns=1.66, num_tools=0.66, failed=0.151, completion_tokens=33.9]
validation: 14%|█▎ | 54/400 [00:13<00:23, 15.03it/s, reward=-0.503, num_turns=1.65, num_tools=0.648, failed=0.167, completion_tokens=33.9]
validation: 14%|█▍ | 55/400 [00:13<00:22, 15.03it/s, reward=-0.548, num_turns=1.64, num_tools=0.636, failed=0.182, completion_tokens=33.9]
validation: 14%|█▍ | 56/400 [00:13<00:22, 15.03it/s, reward=-0.574, num_turns=1.64, num_tools=0.643, failed=0.179, completion_tokens=33.9]
validation: 14%|█▍ | 57/400 [00:13<00:22, 15.03it/s, reward=-0.494, num_turns=1.65, num_tools=0.649, failed=0.175, completion_tokens=33.8]
validation: 14%|█▍ | 58/400 [00:13<00:22, 15.03it/s, reward=-0.46, num_turns=1.66, num_tools=0.655, failed=0.172, completion_tokens=33.5]
validation: 15%|█▍ | 59/400 [00:13<00:22, 15.03it/s, reward=-0.418, num_turns=1.66, num_tools=0.661, failed=0.169, completion_tokens=33.4]
validation: 15%|█▌ | 60/400 [00:13<00:22, 15.03it/s, reward=-0.461, num_turns=1.65, num_tools=0.65, failed=0.167, completion_tokens=35.8]
validation: 15%|█▌ | 61/400 [00:13<00:22, 15.03it/s, reward=-0.486, num_turns=1.66, num_tools=0.656, failed=0.164, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 16%|█▌ | 62/400 [00:13<00:22, 15.03it/s, reward=-0.527, num_turns=1.65, num_tools=0.645, failed=0.177, completion_tokens=35.7]
validation: 16%|█▌ | 63/400 [00:13<00:22, 15.03it/s, reward=-0.566, num_turns=1.63, num_tools=0.635, failed=0.19, completion_tokens=35.7]
validation: 16%|█▌ | 64/400 [00:13<00:22, 15.03it/s, reward=-0.604, num_turns=1.62, num_tools=0.625, failed=0.203, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 16%|█▋ | 65/400 [00:13<00:22, 15.03it/s, reward=-0.641, num_turns=1.62, num_tools=0.615, failed=0.215, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 16%|█▋ | 66/400 [00:13<00:22, 15.03it/s, reward=-0.677, num_turns=1.61, num_tools=0.606, failed=0.227, completion_tokens=35.7]
validation: 17%|█▋ | 67/400 [00:13<00:22, 15.03it/s, reward=-0.711, num_turns=1.6, num_tools=0.597, failed=0.239, completion_tokens=35.7] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 17%|█▋ | 68/400 [00:13<00:22, 15.03it/s, reward=-0.745, num_turns=1.59, num_tools=0.588, failed=0.25, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 17%|█▋ | 69/400 [00:13<00:22, 15.03it/s, reward=-0.778, num_turns=1.58, num_tools=0.58, failed=0.261, completion_tokens=35.7]
validation: 18%|█▊ | 70/400 [00:13<00:21, 15.03it/s, reward=-0.81, num_turns=1.57, num_tools=0.571, failed=0.271, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 18%|█▊ | 71/400 [00:13<00:21, 15.03it/s, reward=-0.84, num_turns=1.56, num_tools=0.563, failed=0.282, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 18%|█▊ | 72/400 [00:13<00:21, 15.03it/s, reward=-0.87, num_turns=1.56, num_tools=0.556, failed=0.292, completion_tokens=35.7]
validation: 18%|█▊ | 73/400 [00:13<00:21, 15.03it/s, reward=-0.9, num_turns=1.55, num_tools=0.548, failed=0.301, completion_tokens=35.7] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 18%|█▊ | 74/400 [00:13<00:21, 15.03it/s, reward=-0.928, num_turns=1.54, num_tools=0.541, failed=0.311, completion_tokens=35.7]
validation: 19%|█▉ | 75/400 [00:13<00:21, 15.03it/s, reward=-0.956, num_turns=1.53, num_tools=0.533, failed=0.32, completion_tokens=35.7] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 19%|█▉ | 76/400 [00:13<00:21, 15.03it/s, reward=-0.982, num_turns=1.53, num_tools=0.526, failed=0.329, completion_tokens=35.7]
validation: 19%|█▉ | 77/400 [00:13<00:21, 15.03it/s, reward=-1.01, num_turns=1.52, num_tools=0.519, failed=0.338, completion_tokens=35.7] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 20%|█▉ | 78/400 [00:13<00:21, 15.03it/s, reward=-1.03, num_turns=1.51, num_tools=0.513, failed=0.346, completion_tokens=35.7]
validation: 20%|█▉ | 79/400 [00:13<00:21, 15.03it/s, reward=-1.06, num_turns=1.51, num_tools=0.506, failed=0.354, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 20%|██ | 80/400 [00:13<00:21, 15.03it/s, reward=-1.08, num_turns=1.5, num_tools=0.5, failed=0.362, completion_tokens=35.7]
validation: 20%|██ | 81/400 [00:13<00:21, 15.03it/s, reward=-1.11, num_turns=1.49, num_tools=0.494, failed=0.37, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 20%|██ | 82/400 [00:13<00:21, 15.03it/s, reward=-1.13, num_turns=1.49, num_tools=0.488, failed=0.378, completion_tokens=35.7]
validation: 21%|██ | 83/400 [00:13<00:21, 15.03it/s, reward=-1.15, num_turns=1.48, num_tools=0.482, failed=0.386, completion_tokens=35.7]
validation: 21%|██ | 84/400 [00:13<00:21, 15.03it/s, reward=-1.17, num_turns=1.48, num_tools=0.476, failed=0.393, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 21%|██▏ | 85/400 [00:13<00:20, 15.03it/s, reward=-1.2, num_turns=1.47, num_tools=0.471, failed=0.4, completion_tokens=35.7]
validation: 22%|██▏ | 86/400 [00:13<00:20, 15.03it/s, reward=-1.22, num_turns=1.47, num_tools=0.465, failed=0.407, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 22%|██▏ | 87/400 [00:13<00:20, 15.03it/s, reward=-1.24, num_turns=1.46, num_tools=0.46, failed=0.414, completion_tokens=35.7]
validation: 22%|██▏ | 88/400 [00:13<00:20, 15.03it/s, reward=-1.26, num_turns=1.45, num_tools=0.455, failed=0.42, completion_tokens=35.7]
validation: 22%|██▏ | 89/400 [00:13<00:20, 15.03it/s, reward=-1.28, num_turns=1.45, num_tools=0.449, failed=0.427, completion_tokens=35.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 22%|██▎ | 90/400 [00:13<00:20, 15.03it/s, reward=-1.3, num_turns=1.44, num_tools=0.444, failed=0.433, completion_tokens=35.7]
validation: 23%|██▎ | 91/400 [00:13<00:20, 15.03it/s, reward=-1.32, num_turns=1.44, num_tools=0.44, failed=0.44, completion_tokens=35.7]
validation: 23%|██▎ | 92/400 [00:13<00:20, 15.03it/s, reward=-1.33, num_turns=1.43, num_tools=0.435, failed=0.446, completion_tokens=35.7]
validation: 23%|██▎ | 93/400 [00:13<00:20, 15.03it/s, reward=-1.35, num_turns=1.43, num_tools=0.43, failed=0.452, completion_tokens=35.7]
validation: 24%|██▎ | 94/400 [00:13<00:20, 15.03it/s, reward=-1.36, num_turns=1.44, num_tools=0.436, failed=0.447, completion_tokens=35.5]
validation: 24%|██▍ | 95/400 [00:13<00:20, 15.03it/s, reward=-1.36, num_turns=1.44, num_tools=0.442, failed=0.442, completion_tokens=35.5]
validation: 24%|██▍ | 96/400 [00:13<00:20, 15.03it/s, reward=-1.34, num_turns=1.45, num_tools=0.448, failed=0.438, completion_tokens=35.2]
validation: 24%|██▍ | 97/400 [00:13<00:20, 15.03it/s, reward=-1.31, num_turns=1.45, num_tools=0.454, failed=0.433, completion_tokens=35]
validation: 24%|██▍ | 98/400 [00:13<00:20, 15.03it/s, reward=-1.28, num_turns=1.46, num_tools=0.459, failed=0.429, completion_tokens=34.8]
validation: 25%|██▍ | 99/400 [00:13<00:20, 15.03it/s, reward=-1.25, num_turns=1.46, num_tools=0.465, failed=0.424, completion_tokens=34.5]
validation: 25%|██▌ | 100/400 [00:13<00:19, 15.03it/s, reward=-1.22, num_turns=1.47, num_tools=0.47, failed=0.42, completion_tokens=34.5]
validation: 25%|██▌ | 101/400 [00:13<00:19, 15.03it/s, reward=-1.23, num_turns=1.48, num_tools=0.475, failed=0.416, completion_tokens=34.4]
validation: 26%|██▌ | 102/400 [00:13<00:06, 47.42it/s, reward=-1.23, num_turns=1.48, num_tools=0.475, failed=0.416, completion_tokens=34.4]
validation: 26%|██▌ | 102/400 [00:13<00:06, 47.42it/s, reward=-1.21, num_turns=1.48, num_tools=0.48, failed=0.412, completion_tokens=34.5]
validation: 26%|██▌ | 103/400 [00:13<00:06, 47.42it/s, reward=-1.22, num_turns=1.49, num_tools=0.485, failed=0.408, completion_tokens=34.4]
validation: 26%|██▌ | 104/400 [00:13<00:06, 47.42it/s, reward=-1.18, num_turns=1.49, num_tools=0.49, failed=0.404, completion_tokens=34.2]
validation: 26%|██▋ | 105/400 [00:13<00:06, 47.42it/s, reward=-1.19, num_turns=1.5, num_tools=0.495, failed=0.4, completion_tokens=34.1]
validation: 26%|██▋ | 106/400 [00:13<00:06, 47.42it/s, reward=-1.2, num_turns=1.5, num_tools=0.5, failed=0.396, completion_tokens=34.2]
validation: 27%|██▋ | 107/400 [00:13<00:06, 47.42it/s, reward=-1.17, num_turns=1.5, num_tools=0.505, failed=0.393, completion_tokens=34.1]
validation: 27%|██▋ | 108/400 [00:13<00:06, 47.42it/s, reward=-1.18, num_turns=1.51, num_tools=0.509, failed=0.389, completion_tokens=33.9]
validation: 27%|██▋ | 109/400 [00:13<00:06, 47.42it/s, reward=-1.16, num_turns=1.51, num_tools=0.514, failed=0.385, completion_tokens=34]
validation: 28%|██▊ | 110/400 [00:13<00:06, 47.42it/s, reward=-1.13, num_turns=1.52, num_tools=0.518, failed=0.382, completion_tokens=33.9]
validation: 28%|██▊ | 111/400 [00:13<00:06, 47.42it/s, reward=-1.14, num_turns=1.52, num_tools=0.523, failed=0.378, completion_tokens=34.1]
validation: 28%|██▊ | 112/400 [00:13<00:06, 47.42it/s, reward=-1.09, num_turns=1.53, num_tools=0.527, failed=0.375, completion_tokens=33.9]
validation: 28%|██▊ | 113/400 [00:13<00:06, 47.42it/s, reward=-1.05, num_turns=1.53, num_tools=0.531, failed=0.372, completion_tokens=33.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 28%|██▊ | 114/400 [00:13<00:06, 47.42it/s, reward=-1.06, num_turns=1.53, num_tools=0.526, failed=0.377, completion_tokens=33.7]
validation: 29%|██▉ | 115/400 [00:13<00:06, 47.42it/s, reward=-1.07, num_turns=1.53, num_tools=0.53, failed=0.374, completion_tokens=33.8] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 29%|██▉ | 116/400 [00:13<00:05, 47.42it/s, reward=-1.09, num_turns=1.53, num_tools=0.526, failed=0.379, completion_tokens=33.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 29%|██▉ | 117/400 [00:13<00:05, 47.42it/s, reward=-1.1, num_turns=1.52, num_tools=0.521, failed=0.385, completion_tokens=33.8] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 30%|██▉ | 118/400 [00:13<00:05, 47.42it/s, reward=-1.12, num_turns=1.52, num_tools=0.517, failed=0.39, completion_tokens=33.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 30%|██▉ | 119/400 [00:13<00:05, 47.42it/s, reward=-1.14, num_turns=1.51, num_tools=0.513, failed=0.395, completion_tokens=33.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 30%|███ | 120/400 [00:13<00:05, 47.42it/s, reward=-1.15, num_turns=1.51, num_tools=0.508, failed=0.4, completion_tokens=33.8] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 30%|███ | 121/400 [00:13<00:04, 57.95it/s, reward=-1.15, num_turns=1.51, num_tools=0.508, failed=0.4, completion_tokens=33.8]
validation: 30%|███ | 121/400 [00:13<00:04, 57.95it/s, reward=-1.17, num_turns=1.5, num_tools=0.504, failed=0.405, completion_tokens=33.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 30%|███ | 122/400 [00:13<00:04, 57.95it/s, reward=-1.18, num_turns=1.5, num_tools=0.5, failed=0.41, completion_tokens=33.8] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 31%|███ | 123/400 [00:13<00:04, 57.95it/s, reward=-1.2, num_turns=1.5, num_tools=0.496, failed=0.415, completion_tokens=33.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 31%|███ | 124/400 [00:13<00:04, 57.95it/s, reward=-1.21, num_turns=1.49, num_tools=0.492, failed=0.419, completion_tokens=33.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 31%|███▏ | 125/400 [00:13<00:04, 57.95it/s, reward=-1.23, num_turns=1.49, num_tools=0.488, failed=0.424, completion_tokens=33.8]
validation: 32%|███▏ | 126/400 [00:13<00:04, 57.95it/s, reward=-1.23, num_turns=1.49, num_tools=0.492, failed=0.421, completion_tokens=34]
validation: 32%|███▏ | 127/400 [00:13<00:04, 57.95it/s, reward=-1.21, num_turns=1.5, num_tools=0.496, failed=0.417, completion_tokens=34.2]
validation: 32%|███▏ | 128/400 [00:13<00:04, 57.95it/s, reward=-1.22, num_turns=1.5, num_tools=0.5, failed=0.414, completion_tokens=34.4]
validation: 32%|███▏ | 129/400 [00:13<00:04, 57.95it/s, reward=-1.22, num_turns=1.5, num_tools=0.504, failed=0.411, completion_tokens=34.5]
validation: 32%|███▎ | 130/400 [00:13<00:04, 57.95it/s, reward=-1.24, num_turns=1.5, num_tools=0.5, failed=0.408, completion_tokens=35.7]
validation: 33%|███▎ | 131/400 [00:13<00:04, 57.95it/s, reward=-1.24, num_turns=1.5, num_tools=0.504, failed=0.405, completion_tokens=36][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 33%|███▎ | 132/400 [00:13<00:04, 57.95it/s, reward=-1.26, num_turns=1.5, num_tools=0.5, failed=0.409, completion_tokens=36]
validation: 33%|███▎ | 133/400 [00:13<00:04, 57.95it/s, reward=-1.24, num_turns=1.5, num_tools=0.504, failed=0.406, completion_tokens=36.4]
validation: 34%|███▎ | 134/400 [00:13<00:04, 57.95it/s, reward=-1.25, num_turns=1.51, num_tools=0.507, failed=0.403, completion_tokens=36.7]
validation: 34%|███▍ | 135/400 [00:13<00:04, 57.95it/s, reward=-1.25, num_turns=1.51, num_tools=0.511, failed=0.4, completion_tokens=37.2]
validation: 34%|███▍ | 136/400 [00:13<00:04, 57.95it/s, reward=-1.26, num_turns=1.51, num_tools=0.515, failed=0.397, completion_tokens=36.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 34%|███▍ | 137/400 [00:13<00:04, 57.95it/s, reward=-1.27, num_turns=1.51, num_tools=0.511, failed=0.401, completion_tokens=36.9]
validation: 34%|███▍ | 138/400 [00:13<00:04, 57.95it/s, reward=-1.28, num_turns=1.51, num_tools=0.507, failed=0.406, completion_tokens=36.9]
validation: 35%|███▍ | 139/400 [00:13<00:04, 54.45it/s, reward=-1.28, num_turns=1.51, num_tools=0.507, failed=0.406, completion_tokens=36.9]
validation: 35%|███▍ | 139/400 [00:13<00:04, 54.45it/s, reward=-1.3, num_turns=1.5, num_tools=0.504, failed=0.403, completion_tokens=36.7]
validation: 35%|███▌ | 140/400 [00:13<00:04, 54.45it/s, reward=-1.26, num_turns=1.51, num_tools=0.507, failed=0.4, completion_tokens=36.6]
validation: 35%|███▌ | 141/400 [00:13<00:04, 54.45it/s, reward=-1.24, num_turns=1.51, num_tools=0.511, failed=0.397, completion_tokens=36.5]
validation: 36%|███▌ | 142/400 [00:13<00:04, 54.45it/s, reward=-1.21, num_turns=1.51, num_tools=0.514, failed=0.394, completion_tokens=36.3]
validation: 36%|███▌ | 143/400 [00:13<00:04, 54.45it/s, reward=-1.19, num_turns=1.52, num_tools=0.517, failed=0.392, completion_tokens=36.1]
validation: 36%|███▌ | 144/400 [00:13<00:04, 54.45it/s, reward=-1.15, num_turns=1.52, num_tools=0.521, failed=0.389, completion_tokens=36]
validation: 36%|███▋ | 145/400 [00:13<00:04, 54.45it/s, reward=-1.12, num_turns=1.52, num_tools=0.524, failed=0.386, completion_tokens=35.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 36%|███▋ | 146/400 [00:13<00:04, 54.45it/s, reward=-1.13, num_turns=1.52, num_tools=0.521, failed=0.39, completion_tokens=35.8]
validation: 37%|███▋ | 147/400 [00:13<00:04, 54.45it/s, reward=-1.12, num_turns=1.52, num_tools=0.524, failed=0.388, completion_tokens=36.3]
validation: 37%|███▋ | 148/400 [00:13<00:04, 54.45it/s, reward=-1.1, num_turns=1.53, num_tools=0.527, failed=0.385, completion_tokens=36.1] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 37%|███▋ | 149/400 [00:14<00:04, 54.45it/s, reward=-1.11, num_turns=1.53, num_tools=0.53, failed=0.389, completion_tokens=36.1]
validation: 38%|███▊ | 150/400 [00:14<00:04, 54.45it/s, reward=-1.12, num_turns=1.53, num_tools=0.533, failed=0.393, completion_tokens=36.1]
validation: 38%|███▊ | 151/400 [00:14<00:04, 54.45it/s, reward=-1.13, num_turns=1.54, num_tools=0.536, failed=0.397, completion_tokens=36.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 38%|███▊ | 152/400 [00:14<00:04, 54.45it/s, reward=-1.15, num_turns=1.54, num_tools=0.539, failed=0.401, completion_tokens=36.2]
validation: 38%|███▊ | 153/400 [00:14<00:04, 54.45it/s, reward=-1.16, num_turns=1.54, num_tools=0.542, failed=0.405, completion_tokens=36]
validation: 38%|███▊ | 154/400 [00:14<00:04, 54.45it/s, reward=-1.17, num_turns=1.55, num_tools=0.545, failed=0.409, completion_tokens=35.9]
validation: 39%|███▉ | 155/400 [00:14<00:04, 54.45it/s, reward=-1.18, num_turns=1.55, num_tools=0.548, failed=0.413, completion_tokens=35.7]
validation: 39%|███▉ | 156/400 [00:14<00:04, 54.45it/s, reward=-1.16, num_turns=1.55, num_tools=0.551, failed=0.41, completion_tokens=35.6]
validation: 39%|███▉ | 157/400 [00:14<00:04, 54.45it/s, reward=-1.14, num_turns=1.55, num_tools=0.554, failed=0.408, completion_tokens=35.4]
validation: 40%|███▉ | 158/400 [00:14<00:04, 54.45it/s, reward=-1.12, num_turns=1.56, num_tools=0.557, failed=0.405, completion_tokens=35.3][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 40%|███▉ | 159/400 [00:14<00:04, 54.45it/s, reward=-1.13, num_turns=1.55, num_tools=0.553, failed=0.409, completion_tokens=35.3]
validation: 40%|████ | 160/400 [00:14<00:04, 54.45it/s, reward=-1.14, num_turns=1.56, num_tools=0.556, failed=0.406, completion_tokens=35.2]
validation: 40%|████ | 161/400 [00:14<00:04, 54.45it/s, reward=-1.12, num_turns=1.56, num_tools=0.559, failed=0.404, completion_tokens=35] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 40%|████ | 162/400 [00:14<00:03, 72.12it/s, reward=-1.12, num_turns=1.56, num_tools=0.559, failed=0.404, completion_tokens=35]
validation: 40%|████ | 162/400 [00:14<00:03, 72.12it/s, reward=-1.13, num_turns=1.56, num_tools=0.562, failed=0.407, completion_tokens=34.9]
validation: 41%|████ | 163/400 [00:14<00:03, 72.12it/s, reward=-1.14, num_turns=1.56, num_tools=0.564, failed=0.411, completion_tokens=34.9]
validation: 41%|████ | 164/400 [00:14<00:03, 72.12it/s, reward=-1.15, num_turns=1.57, num_tools=0.567, failed=0.409, completion_tokens=34.8]
validation: 41%|████▏ | 165/400 [00:14<00:03, 72.12it/s, reward=-1.15, num_turns=1.57, num_tools=0.57, failed=0.406, completion_tokens=34.7]
validation: 42%|████▏ | 166/400 [00:14<00:03, 72.12it/s, reward=-1.16, num_turns=1.57, num_tools=0.572, failed=0.404, completion_tokens=35]
validation: 42%|████▏ | 167/400 [00:14<00:03, 72.12it/s, reward=-1.14, num_turns=1.57, num_tools=0.575, failed=0.401, completion_tokens=35]
validation: 42%|████▏ | 168/400 [00:14<00:03, 72.12it/s, reward=-1.11, num_turns=1.58, num_tools=0.577, failed=0.399, completion_tokens=34.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 42%|████▏ | 169/400 [00:14<00:03, 72.12it/s, reward=-1.13, num_turns=1.57, num_tools=0.574, failed=0.402, completion_tokens=34.8]
validation: 42%|████▎ | 170/400 [00:14<00:03, 72.12it/s, reward=-1.13, num_turns=1.58, num_tools=0.576, failed=0.4, completion_tokens=34.8]
validation: 43%|████▎ | 171/400 [00:14<00:03, 72.12it/s, reward=-1.14, num_turns=1.58, num_tools=0.579, failed=0.398, completion_tokens=34.8][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 43%|████▎ | 172/400 [00:14<00:03, 72.12it/s, reward=-1.15, num_turns=1.58, num_tools=0.581, failed=0.401, completion_tokens=34.7]
validation: 43%|████▎ | 173/400 [00:14<00:03, 72.12it/s, reward=-1.12, num_turns=1.58, num_tools=0.584, failed=0.399, completion_tokens=34.6]
validation: 44%|████▎ | 174/400 [00:14<00:03, 72.12it/s, reward=-1.09, num_turns=1.59, num_tools=0.586, failed=0.397, completion_tokens=34.5]
validation: 44%|████▍ | 175/400 [00:14<00:03, 72.12it/s, reward=-1.06, num_turns=1.59, num_tools=0.589, failed=0.394, completion_tokens=34.4]
validation: 44%|████▍ | 176/400 [00:14<00:03, 72.12it/s, reward=-1.06, num_turns=1.59, num_tools=0.591, failed=0.392, completion_tokens=34.4]
validation: 44%|████▍ | 177/400 [00:14<00:03, 72.12it/s, reward=-1.03, num_turns=1.59, num_tools=0.593, failed=0.39, completion_tokens=34.3]
validation: 44%|████▍ | 178/400 [00:14<00:03, 72.12it/s, reward=-1.01, num_turns=1.6, num_tools=0.596, failed=0.388, completion_tokens=34.3]
validation: 45%|████▍ | 179/400 [00:14<00:02, 76.04it/s, reward=-1.01, num_turns=1.6, num_tools=0.596, failed=0.388, completion_tokens=34.3]
validation: 45%|████▍ | 179/400 [00:14<00:02, 76.04it/s, reward=-0.979, num_turns=1.6, num_tools=0.598, failed=0.385, completion_tokens=34.2]
validation: 45%|████▌ | 180/400 [00:14<00:02, 76.04it/s, reward=-0.951, num_turns=1.6, num_tools=0.6, failed=0.383, completion_tokens=34.2]
validation: 45%|████▌ | 181/400 [00:14<00:02, 76.04it/s, reward=-0.924, num_turns=1.6, num_tools=0.602, failed=0.381, completion_tokens=34.1]
validation: 46%|████▌ | 182/400 [00:14<00:02, 76.04it/s, reward=-0.897, num_turns=1.6, num_tools=0.604, failed=0.379, completion_tokens=34.1]
validation: 46%|████▌ | 183/400 [00:14<00:02, 76.04it/s, reward=-0.87, num_turns=1.61, num_tools=0.607, failed=0.377, completion_tokens=34]
validation: 46%|████▌ | 184/400 [00:14<00:02, 76.04it/s, reward=-0.862, num_turns=1.61, num_tools=0.609, failed=0.375, completion_tokens=34]
validation: 46%|████▋ | 185/400 [00:14<00:02, 76.04it/s, reward=-0.855, num_turns=1.61, num_tools=0.611, failed=0.373, completion_tokens=33.9]
validation: 46%|████▋ | 186/400 [00:14<00:02, 76.04it/s, reward=-0.861, num_turns=1.61, num_tools=0.613, failed=0.371, completion_tokens=33.9]
validation: 47%|████▋ | 187/400 [00:14<00:02, 76.04it/s, reward=-0.867, num_turns=1.61, num_tools=0.615, failed=0.369, completion_tokens=33.8]
validation: 47%|████▋ | 188/400 [00:14<00:02, 76.04it/s, reward=-0.873, num_turns=1.62, num_tools=0.617, failed=0.367, completion_tokens=33.7]
validation: 47%|████▋ | 189/400 [00:14<00:02, 76.04it/s, reward=-0.853, num_turns=1.62, num_tools=0.619, failed=0.365, completion_tokens=33.7]
validation: 48%|████▊ | 190/400 [00:14<00:02, 76.04it/s, reward=-0.859, num_turns=1.62, num_tools=0.621, failed=0.363, completion_tokens=33.7]
validation: 48%|████▊ | 191/400 [00:14<00:02, 76.04it/s, reward=-0.865, num_turns=1.62, num_tools=0.623, failed=0.361, completion_tokens=33.6]
validation: 48%|████▊ | 192/400 [00:14<00:02, 76.04it/s, reward=-0.85, num_turns=1.62, num_tools=0.625, failed=0.359, completion_tokens=33.6]
validation: 48%|████▊ | 193/400 [00:14<00:02, 76.04it/s, reward=-0.835, num_turns=1.63, num_tools=0.627, failed=0.358, completion_tokens=33.5]
validation: 48%|████▊ | 194/400 [00:14<00:02, 76.04it/s, reward=-0.828, num_turns=1.63, num_tools=0.629, failed=0.356, completion_tokens=33.5]
validation: 49%|████▉ | 195/400 [00:14<00:02, 76.04it/s, reward=-0.834, num_turns=1.63, num_tools=0.631, failed=0.354, completion_tokens=33.4]
validation: 49%|████▉ | 196/400 [00:14<00:02, 76.04it/s, reward=-0.84, num_turns=1.63, num_tools=0.633, failed=0.352, completion_tokens=33.5]
validation: 49%|████▉ | 197/400 [00:14<00:02, 76.04it/s, reward=-0.846, num_turns=1.63, num_tools=0.635, failed=0.35, completion_tokens=33.5]
validation: 50%|████▉ | 198/400 [00:14<00:02, 76.04it/s, reward=-0.832, num_turns=1.64, num_tools=0.636, failed=0.348, completion_tokens=33.5]
validation: 50%|████▉ | 199/400 [00:14<00:02, 92.39it/s, reward=-0.832, num_turns=1.64, num_tools=0.636, failed=0.348, completion_tokens=33.5]
validation: 50%|████▉ | 199/400 [00:14<00:02, 92.39it/s, reward=-0.838, num_turns=1.64, num_tools=0.638, failed=0.347, completion_tokens=33.4]
validation: 50%|█████ | 200/400 [00:14<00:02, 92.39it/s, reward=-0.821, num_turns=1.64, num_tools=0.64, failed=0.345, completion_tokens=33.5]
validation: 50%|█████ | 201/400 [00:14<00:02, 92.39it/s, reward=-0.827, num_turns=1.64, num_tools=0.642, failed=0.343, completion_tokens=33.4]
validation: 50%|█████ | 202/400 [00:14<00:02, 92.39it/s, reward=-0.833, num_turns=1.64, num_tools=0.644, failed=0.342, completion_tokens=33.4]
validation: 51%|█████ | 203/400 [00:14<00:02, 92.39it/s, reward=-0.838, num_turns=1.65, num_tools=0.645, failed=0.34, completion_tokens=33.8]
validation: 51%|█████ | 204/400 [00:14<00:02, 92.39it/s, reward=-0.844, num_turns=1.65, num_tools=0.647, failed=0.338, completion_tokens=33.8]
validation: 51%|█████▏ | 205/400 [00:14<00:02, 92.39it/s, reward=-0.85, num_turns=1.65, num_tools=0.654, failed=0.337, completion_tokens=33.9]
validation: 52%|█████▏ | 206/400 [00:14<00:02, 92.39it/s, reward=-0.855, num_turns=1.66, num_tools=0.655, failed=0.335, completion_tokens=33.9]
validation: 52%|█████▏ | 207/400 [00:14<00:02, 92.39it/s, reward=-0.841, num_turns=1.66, num_tools=0.657, failed=0.333, completion_tokens=33.8]
validation: 52%|█████▏ | 208/400 [00:14<00:02, 92.39it/s, reward=-0.828, num_turns=1.66, num_tools=0.659, failed=0.332, completion_tokens=33.8]
validation: 52%|█████▏ | 209/400 [00:14<00:02, 92.39it/s, reward=-0.833, num_turns=1.66, num_tools=0.66, failed=0.33, completion_tokens=33.8]
validation: 52%|█████▎ | 210/400 [00:14<00:02, 92.39it/s, reward=-0.839, num_turns=1.66, num_tools=0.662, failed=0.329, completion_tokens=33.8]
validation: 53%|█████▎ | 211/400 [00:14<00:02, 92.39it/s, reward=-0.823, num_turns=1.66, num_tools=0.664, failed=0.327, completion_tokens=33.9]
validation: 53%|█████▎ | 212/400 [00:14<00:02, 92.39it/s, reward=-0.829, num_turns=1.67, num_tools=0.665, failed=0.325, completion_tokens=33.9]
validation: 53%|█████▎ | 213/400 [00:14<00:02, 92.39it/s, reward=-0.834, num_turns=1.67, num_tools=0.667, failed=0.324, completion_tokens=33.9]
validation: 54%|█████▎ | 214/400 [00:14<00:02, 92.39it/s, reward=-0.819, num_turns=1.67, num_tools=0.668, failed=0.322, completion_tokens=33.8]
validation: 54%|█████▍ | 215/400 [00:14<00:02, 86.87it/s, reward=-0.819, num_turns=1.67, num_tools=0.668, failed=0.322, completion_tokens=33.8]
validation: 54%|█████▍ | 215/400 [00:14<00:02, 86.87it/s, reward=-0.808, num_turns=1.67, num_tools=0.67, failed=0.321, completion_tokens=33.9]
validation: 54%|█████▍ | 216/400 [00:14<00:02, 86.87it/s, reward=-0.795, num_turns=1.67, num_tools=0.671, failed=0.319, completion_tokens=33.9]
validation: 54%|█████▍ | 217/400 [00:14<00:02, 86.87it/s, reward=-0.8, num_turns=1.67, num_tools=0.673, failed=0.318, completion_tokens=33.9]
validation: 55%|█████▍ | 218/400 [00:14<00:02, 86.87it/s, reward=-0.806, num_turns=1.67, num_tools=0.674, failed=0.317, completion_tokens=33.8]
validation: 55%|█████▍ | 219/400 [00:14<00:02, 86.87it/s, reward=-0.811, num_turns=1.68, num_tools=0.676, failed=0.315, completion_tokens=34]
validation: 55%|█████▌ | 220/400 [00:14<00:02, 86.87it/s, reward=-0.817, num_turns=1.68, num_tools=0.677, failed=0.314, completion_tokens=34]
validation: 55%|█████▌ | 221/400 [00:14<00:02, 86.87it/s, reward=-0.822, num_turns=1.68, num_tools=0.679, failed=0.312, completion_tokens=33.9]
validation: 56%|█████▌ | 222/400 [00:14<00:02, 86.87it/s, reward=-0.827, num_turns=1.68, num_tools=0.685, failed=0.311, completion_tokens=34]
validation: 56%|█████▌ | 223/400 [00:14<00:02, 86.87it/s, reward=-0.833, num_turns=1.69, num_tools=0.691, failed=0.309, completion_tokens=34.2]
validation: 56%|█████▌ | 224/400 [00:14<00:02, 86.87it/s, reward=-0.818, num_turns=1.69, num_tools=0.692, failed=0.308, completion_tokens=34.5]
validation: 56%|█████▋ | 225/400 [00:14<00:02, 86.87it/s, reward=-0.827, num_turns=1.69, num_tools=0.689, failed=0.307, completion_tokens=35]
validation: 56%|█████▋ | 226/400 [00:14<00:02, 86.87it/s, reward=-0.813, num_turns=1.69, num_tools=0.69, failed=0.305, completion_tokens=34.9]
validation: 57%|█████▋ | 227/400 [00:14<00:01, 86.87it/s, reward=-0.818, num_turns=1.69, num_tools=0.692, failed=0.304, completion_tokens=34.9]
validation: 57%|█████▋ | 228/400 [00:14<00:01, 86.87it/s, reward=-0.827, num_turns=1.69, num_tools=0.689, failed=0.303, completion_tokens=37.7][2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.473315 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.395957 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.458265 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.430569 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.428553 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.450014 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.401053 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.434205 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.475570 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.407062 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.430383 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.468651 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.404790 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.422452 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.411601 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.474694 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.486884 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.488547 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.428166 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.414558 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.447537 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.453323 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.436907 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.446272 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.496000 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.453066 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.488989 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.478658 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.451624 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.432928 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.420997 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.474647 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.447945 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.449846 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.468465 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.448000 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.409914 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.427265 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.383793 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.403102 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.429834 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.378778 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.444976 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.380470 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.484490 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.379489 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.488917 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.443552 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.484126 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.377713 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.455466 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.442418 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.411960 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.478103 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.397940 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.386468 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.433327 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.437740 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.415018 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.457598 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.395625 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.382885 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.414943 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.408685 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.467779 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.414284 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.449459 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.437600 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.484733 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.490503 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.491294 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.492965 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.478945 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.448204 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.479687 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.458371 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.452620 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.416544 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.466102 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.443001 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.438199 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.386365 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.420463 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.398959 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.415648 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.406742 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.422783 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.468912 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.457281 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.447545 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.404803 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.380038 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.377342 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.471050 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.416476 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.473025 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.460678 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.408835 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.444229 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.387272 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.482396 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.418123 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.453410 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.462687 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.413345 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.438264 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.383264 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.388137 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.494660 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.470784 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.495510 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.438529 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.399639 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.396314 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.392011 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.414534 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.497475 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.393452 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.426093 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.442541 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.382879 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.406868 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.442518 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.409651 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.452520 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.440985 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.396273 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.440177 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.483216 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.488335 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.379325 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.404576 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.411981 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.418952 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.475371 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.477682 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.379507 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.487781 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.404374 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.385444 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.377447 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.456144 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.411518 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.472613 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.448824 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.418550 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.427137 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.466364 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.481079 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.424927 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.385840 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.434185 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.423498 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.470993 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.421335 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.445158 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.401004 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.402147 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.443400 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.427557 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.438579 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.475885 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.469949 seconds
[2026-04-13 02:44:50] INFO _base_client.py:1693: Retrying request to /chat/completions in 0.483275 seconds
validation: 57%|█████▋ | 229/400 [00:15<00:03, 54.32it/s, reward=-0.827, num_turns=1.69, num_tools=0.689, failed=0.303, completion_tokens=37.7]
validation: 57%|█████▋ | 229/400 [00:15<00:03, 54.32it/s, reward=-0.813, num_turns=1.69, num_tools=0.69, failed=0.301, completion_tokens=38]
validation: 57%|█████▊ | 230/400 [00:15<00:03, 54.32it/s, reward=-0.822, num_turns=1.69, num_tools=0.687, failed=0.3, completion_tokens=40.4]
validation: 58%|█████▊ | 231/400 [00:15<00:03, 54.32it/s, reward=-0.828, num_turns=1.69, num_tools=0.693, failed=0.299, completion_tokens=40.4]
validation: 58%|█████▊ | 232/400 [00:15<00:03, 54.32it/s, reward=-0.813, num_turns=1.69, num_tools=0.694, failed=0.297, completion_tokens=40.7]
validation: 58%|█████▊ | 233/400 [00:15<00:03, 54.32it/s, reward=-0.818, num_turns=1.7, num_tools=0.695, failed=0.296, completion_tokens=41.2]
validation: 58%|█████▊ | 234/400 [00:15<00:03, 54.32it/s, reward=-0.828, num_turns=1.69, num_tools=0.692, failed=0.295, completion_tokens=43.7]
validation: 59%|█████▉ | 235/400 [00:17<00:03, 54.32it/s, reward=-0.837, num_turns=1.69, num_tools=0.689, failed=0.294, completion_tokens=43.6]
validation: 59%|█████▉ | 236/400 [00:17<00:03, 54.32it/s, reward=-0.846, num_turns=1.69, num_tools=0.686, failed=0.292, completion_tokens=43.6]
validation: 59%|█████▉ | 237/400 [00:17<00:03, 54.32it/s, reward=-0.851, num_turns=1.69, num_tools=0.688, failed=0.291, completion_tokens=43.6]
validation: 60%|█████▉ | 238/400 [00:17<00:02, 54.32it/s, reward=-0.86, num_turns=1.68, num_tools=0.685, failed=0.29, completion_tokens=43.7]
validation: 60%|█████▉ | 239/400 [00:17<00:10, 15.66it/s, reward=-0.86, num_turns=1.68, num_tools=0.685, failed=0.29, completion_tokens=43.7]
validation: 60%|█████▉ | 239/400 [00:17<00:10, 15.66it/s, reward=-0.869, num_turns=1.68, num_tools=0.682, failed=0.289, completion_tokens=44.4]
validation: 60%|██████ | 240/400 [00:17<00:10, 15.66it/s, reward=-0.878, num_turns=1.68, num_tools=0.679, failed=0.287, completion_tokens=44.4]
validation: 60%|██████ | 241/400 [00:17<00:10, 15.66it/s, reward=-0.887, num_turns=1.68, num_tools=0.676, failed=0.286, completion_tokens=46.9]
validation: 60%|██████ | 242/400 [00:17<00:10, 15.66it/s, reward=-0.895, num_turns=1.67, num_tools=0.674, failed=0.285, completion_tokens=47.1]
validation: 61%|██████ | 243/400 [00:17<00:10, 15.66it/s, reward=-0.904, num_turns=1.67, num_tools=0.671, failed=0.284, completion_tokens=47.1]
validation: 61%|██████ | 244/400 [00:17<00:09, 15.66it/s, reward=-0.892, num_turns=1.67, num_tools=0.672, failed=0.283, completion_tokens=46.9]
validation: 61%|██████▏ | 245/400 [00:17<00:09, 15.66it/s, reward=-0.897, num_turns=1.67, num_tools=0.673, failed=0.282, completion_tokens=46.8]
validation: 62%|██████▏ | 246/400 [00:17<00:09, 16.76it/s, reward=-0.897, num_turns=1.67, num_tools=0.673, failed=0.282, completion_tokens=46.8]
validation: 62%|██████▏ | 246/400 [00:17<00:09, 16.76it/s, reward=-0.877, num_turns=1.67, num_tools=0.675, failed=0.28, completion_tokens=46.7]
validation: 62%|██████▏ | 247/400 [00:17<00:09, 16.76it/s, reward=-0.881, num_turns=1.68, num_tools=0.676, failed=0.279, completion_tokens=46.5]
validation: 62%|██████▏ | 248/400 [00:17<00:09, 16.76it/s, reward=-0.89, num_turns=1.67, num_tools=0.673, failed=0.278, completion_tokens=48] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 62%|██████▏ | 249/400 [00:17<00:09, 16.76it/s, reward=-0.898, num_turns=1.67, num_tools=0.671, failed=0.281, completion_tokens=48]
validation: 62%|██████▎ | 250/400 [00:18<00:08, 16.76it/s, reward=-0.903, num_turns=1.67, num_tools=0.672, failed=0.28, completion_tokens=47.9]
validation: 63%|██████▎ | 251/400 [00:18<00:08, 16.76it/s, reward=-0.907, num_turns=1.67, num_tools=0.673, failed=0.279, completion_tokens=47.7]
validation: 63%|██████▎ | 252/400 [00:18<00:08, 18.18it/s, reward=-0.907, num_turns=1.67, num_tools=0.673, failed=0.279, completion_tokens=47.7]
validation: 63%|██████▎ | 252/400 [00:18<00:08, 18.18it/s, reward=-0.911, num_turns=1.67, num_tools=0.675, failed=0.278, completion_tokens=47.6]
validation: 63%|██████▎ | 253/400 [00:18<00:08, 18.18it/s, reward=-0.916, num_turns=1.68, num_tools=0.676, failed=0.277, completion_tokens=47.4]
validation: 64%|██████▎ | 254/400 [00:18<00:08, 18.18it/s, reward=-0.92, num_turns=1.68, num_tools=0.677, failed=0.276, completion_tokens=47.3]
validation: 64%|██████▍ | 255/400 [00:18<00:07, 18.18it/s, reward=-0.924, num_turns=1.68, num_tools=0.678, failed=0.275, completion_tokens=47.1]
validation: 64%|██████▍ | 256/400 [00:18<00:07, 18.18it/s, reward=-0.928, num_turns=1.68, num_tools=0.68, failed=0.273, completion_tokens=47]
validation: 64%|██████▍ | 257/400 [00:18<00:08, 17.78it/s, reward=-0.928, num_turns=1.68, num_tools=0.68, failed=0.273, completion_tokens=47]
validation: 64%|██████▍ | 257/400 [00:18<00:08, 17.78it/s, reward=-0.933, num_turns=1.68, num_tools=0.681, failed=0.272, completion_tokens=46.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 64%|██████▍ | 258/400 [00:18<00:07, 17.78it/s, reward=-0.941, num_turns=1.68, num_tools=0.678, failed=0.275, completion_tokens=46.9]
validation: 65%|██████▍ | 259/400 [00:18<00:07, 17.78it/s, reward=-0.949, num_turns=1.68, num_tools=0.676, failed=0.278, completion_tokens=46.9]
validation: 65%|██████▌ | 260/400 [00:18<00:07, 17.78it/s, reward=-0.956, num_turns=1.67, num_tools=0.673, failed=0.281, completion_tokens=46.9]
validation: 65%|██████▌ | 261/400 [00:18<00:07, 17.78it/s, reward=-0.937, num_turns=1.67, num_tools=0.674, failed=0.28, completion_tokens=46.9]
validation: 66%|██████▌ | 262/400 [00:18<00:07, 17.78it/s, reward=-0.919, num_turns=1.68, num_tools=0.676, failed=0.279, completion_tokens=46.8]
validation: 66%|██████▌ | 263/400 [00:18<00:07, 17.78it/s, reward=-0.923, num_turns=1.68, num_tools=0.677, failed=0.278, completion_tokens=46.7]
validation: 66%|██████▌ | 264/400 [00:18<00:07, 17.78it/s, reward=-0.917, num_turns=1.68, num_tools=0.678, failed=0.277, completion_tokens=46.6]
validation: 66%|██████▋ | 265/400 [00:18<00:07, 17.78it/s, reward=-0.908, num_turns=1.68, num_tools=0.679, failed=0.275, completion_tokens=46.6]
validation: 66%|██████▋ | 266/400 [00:18<00:07, 17.78it/s, reward=-0.897, num_turns=1.68, num_tools=0.68, failed=0.274, completion_tokens=46.5]
validation: 67%|██████▋ | 267/400 [00:18<00:07, 17.78it/s, reward=-0.879, num_turns=1.68, num_tools=0.682, failed=0.273, completion_tokens=46.3]
validation: 67%|██████▋ | 268/400 [00:18<00:07, 17.78it/s, reward=-0.861, num_turns=1.68, num_tools=0.683, failed=0.272, completion_tokens=46.2]
validation: 67%|██████▋ | 269/400 [00:18<00:07, 17.78it/s, reward=-0.854, num_turns=1.68, num_tools=0.684, failed=0.271, completion_tokens=46.1]
validation: 68%|██████▊ | 270/400 [00:18<00:07, 17.78it/s, reward=-0.858, num_turns=1.69, num_tools=0.685, failed=0.27, completion_tokens=46]
validation: 68%|██████▊ | 271/400 [00:18<00:07, 17.78it/s, reward=-0.862, num_turns=1.69, num_tools=0.686, failed=0.269, completion_tokens=45.9]
validation: 68%|██████▊ | 272/400 [00:18<00:07, 17.78it/s, reward=-0.85, num_turns=1.69, num_tools=0.688, failed=0.268, completion_tokens=45.8]
validation: 68%|██████▊ | 273/400 [00:18<00:07, 17.78it/s, reward=-0.854, num_turns=1.69, num_tools=0.689, failed=0.267, completion_tokens=45.7]
validation: 68%|██████▊ | 274/400 [00:18<00:07, 17.78it/s, reward=-0.858, num_turns=1.69, num_tools=0.69, failed=0.266, completion_tokens=45.6]
validation: 69%|██████▉ | 275/400 [00:18<00:07, 17.78it/s, reward=-0.841, num_turns=1.69, num_tools=0.691, failed=0.265, completion_tokens=45.5]
validation: 69%|██████▉ | 276/400 [00:18<00:06, 17.78it/s, reward=-0.845, num_turns=1.69, num_tools=0.692, failed=0.264, completion_tokens=45.4]
validation: 69%|██████▉ | 277/400 [00:18<00:06, 17.78it/s, reward=-0.827, num_turns=1.69, num_tools=0.693, failed=0.264, completion_tokens=45.3]
validation: 70%|██████▉ | 278/400 [00:18<00:06, 17.78it/s, reward=-0.81, num_turns=1.69, num_tools=0.694, failed=0.263, completion_tokens=45.2]
validation: 70%|██████▉ | 279/400 [00:18<00:06, 17.78it/s, reward=-0.793, num_turns=1.7, num_tools=0.695, failed=0.262, completion_tokens=45.2]
validation: 70%|███████ | 280/400 [00:18<00:06, 17.78it/s, reward=-0.776, num_turns=1.7, num_tools=0.696, failed=0.261, completion_tokens=45.1]
validation: 70%|███████ | 281/400 [00:18<00:06, 17.78it/s, reward=-0.759, num_turns=1.7, num_tools=0.698, failed=0.26, completion_tokens=45]
validation: 70%|███████ | 282/400 [00:18<00:06, 17.78it/s, reward=-0.742, num_turns=1.7, num_tools=0.699, failed=0.259, completion_tokens=44.9]
validation: 71%|███████ | 283/400 [00:18<00:06, 17.78it/s, reward=-0.725, num_turns=1.7, num_tools=0.7, failed=0.258, completion_tokens=44.9]
validation: 71%|███████ | 284/400 [00:18<00:06, 17.78it/s, reward=-0.708, num_turns=1.7, num_tools=0.701, failed=0.257, completion_tokens=44.8]
validation: 71%|███████▏ | 285/400 [00:18<00:06, 17.78it/s, reward=-0.692, num_turns=1.7, num_tools=0.702, failed=0.256, completion_tokens=44.7]
validation: 72%|███████▏ | 286/400 [00:18<00:06, 17.78it/s, reward=-0.675, num_turns=1.7, num_tools=0.703, failed=0.255, completion_tokens=44.6]
validation: 72%|███████▏ | 287/400 [00:18<00:06, 17.78it/s, reward=-0.659, num_turns=1.7, num_tools=0.704, failed=0.254, completion_tokens=44.5]
validation: 72%|███████▏ | 288/400 [00:18<00:06, 17.78it/s, reward=-0.653, num_turns=1.7, num_tools=0.705, failed=0.253, completion_tokens=44.5]
validation: 72%|███████▏ | 289/400 [00:18<00:06, 17.78it/s, reward=-0.648, num_turns=1.71, num_tools=0.706, failed=0.253, completion_tokens=44.4][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 72%|███████▎ | 290/400 [00:18<00:06, 17.78it/s, reward=-0.656, num_turns=1.7, num_tools=0.703, failed=0.255, completion_tokens=44.4] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 73%|███████▎ | 291/400 [00:18<00:06, 17.78it/s, reward=-0.664, num_turns=1.7, num_tools=0.701, failed=0.258, completion_tokens=44.4][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 73%|███████▎ | 292/400 [00:18<00:06, 17.78it/s, reward=-0.672, num_turns=1.7, num_tools=0.699, failed=0.26, completion_tokens=44.4] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 73%|███████▎ | 293/400 [00:18<00:02, 44.50it/s, reward=-0.672, num_turns=1.7, num_tools=0.699, failed=0.26, completion_tokens=44.4]
validation: 73%|███████▎ | 293/400 [00:18<00:02, 44.50it/s, reward=-0.68, num_turns=1.7, num_tools=0.696, failed=0.263, completion_tokens=44.4]
validation: 74%|███████▎ | 294/400 [00:18<00:02, 44.50it/s, reward=-0.684, num_turns=1.7, num_tools=0.697, failed=0.262, completion_tokens=44.3]
validation: 74%|███████▍ | 295/400 [00:18<00:02, 44.50it/s, reward=-0.689, num_turns=1.7, num_tools=0.698, failed=0.261, completion_tokens=44.3]
validation: 74%|███████▍ | 296/400 [00:18<00:02, 44.50it/s, reward=-0.693, num_turns=1.7, num_tools=0.699, failed=0.26, completion_tokens=44.2]
validation: 74%|███████▍ | 297/400 [00:18<00:02, 44.50it/s, reward=-0.687, num_turns=1.7, num_tools=0.7, failed=0.259, completion_tokens=44.2]
validation: 74%|███████▍ | 298/400 [00:18<00:02, 44.50it/s, reward=-0.692, num_turns=1.7, num_tools=0.701, failed=0.258, completion_tokens=44.2]
validation: 75%|███████▍ | 299/400 [00:18<00:02, 44.50it/s, reward=-0.696, num_turns=1.7, num_tools=0.702, failed=0.258, completion_tokens=44.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 75%|███████▌ | 300/400 [00:18<00:02, 44.50it/s, reward=-0.704, num_turns=1.7, num_tools=0.703, failed=0.26, completion_tokens=44.1] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 75%|███████▌ | 301/400 [00:18<00:02, 44.50it/s, reward=-0.712, num_turns=1.7, num_tools=0.704, failed=0.262, completion_tokens=44.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 76%|███████▌ | 302/400 [00:18<00:02, 44.50it/s, reward=-0.719, num_turns=1.71, num_tools=0.705, failed=0.265, completion_tokens=44]
validation: 76%|███████▌ | 303/400 [00:18<00:02, 44.50it/s, reward=-0.727, num_turns=1.71, num_tools=0.706, failed=0.267, completion_tokens=44][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 76%|███████▌ | 304/400 [00:18<00:02, 44.50it/s, reward=-0.734, num_turns=1.71, num_tools=0.707, failed=0.27, completion_tokens=44] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 76%|███████▋ | 305/400 [00:18<00:02, 44.50it/s, reward=-0.742, num_turns=1.7, num_tools=0.705, failed=0.272, completion_tokens=44]
validation: 76%|███████▋ | 306/400 [00:18<00:02, 44.50it/s, reward=-0.749, num_turns=1.71, num_tools=0.706, failed=0.275, completion_tokens=43.9][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 77%|███████▋ | 307/400 [00:18<00:01, 53.12it/s, reward=-0.749, num_turns=1.71, num_tools=0.706, failed=0.275, completion_tokens=43.9]
validation: 77%|███████▋ | 307/400 [00:18<00:01, 53.12it/s, reward=-0.756, num_turns=1.71, num_tools=0.707, failed=0.277, completion_tokens=43.9]
validation: 77%|███████▋ | 308/400 [00:18<00:01, 53.12it/s, reward=-0.764, num_turns=1.7, num_tools=0.705, failed=0.279, completion_tokens=43.9]
validation: 77%|███████▋ | 309/400 [00:18<00:01, 53.12it/s, reward=-0.756, num_turns=1.71, num_tools=0.706, failed=0.278, completion_tokens=43.9]
validation: 78%|███████▊ | 310/400 [00:18<00:01, 53.12it/s, reward=-0.76, num_turns=1.71, num_tools=0.706, failed=0.277, completion_tokens=43.9]
validation: 78%|███████▊ | 311/400 [00:18<00:01, 53.12it/s, reward=-0.75, num_turns=1.71, num_tools=0.707, failed=0.277, completion_tokens=43.8]
validation: 78%|███████▊ | 312/400 [00:18<00:01, 53.12it/s, reward=-0.743, num_turns=1.71, num_tools=0.708, failed=0.276, completion_tokens=43.9]
validation: 78%|███████▊ | 313/400 [00:18<00:01, 53.12it/s, reward=-0.728, num_turns=1.71, num_tools=0.709, failed=0.275, completion_tokens=43.8]
validation: 78%|███████▊ | 314/400 [00:18<00:01, 53.12it/s, reward=-0.718, num_turns=1.71, num_tools=0.71, failed=0.274, completion_tokens=43.8]
validation: 79%|███████▉ | 315/400 [00:18<00:01, 53.12it/s, reward=-0.722, num_turns=1.71, num_tools=0.711, failed=0.273, completion_tokens=43.8]
validation: 79%|███████▉ | 316/400 [00:18<00:01, 53.12it/s, reward=-0.716, num_turns=1.71, num_tools=0.712, failed=0.272, completion_tokens=43.7]
validation: 79%|███████▉ | 317/400 [00:18<00:01, 53.12it/s, reward=-0.708, num_turns=1.71, num_tools=0.713, failed=0.271, completion_tokens=43.6]
validation: 80%|███████▉ | 318/400 [00:18<00:01, 53.12it/s, reward=-0.699, num_turns=1.71, num_tools=0.714, failed=0.27, completion_tokens=43.5]
validation: 80%|███████▉ | 319/400 [00:18<00:01, 53.12it/s, reward=-0.691, num_turns=1.71, num_tools=0.715, failed=0.27, completion_tokens=43.5]
validation: 80%|████████ | 320/400 [00:18<00:01, 47.62it/s, reward=-0.691, num_turns=1.71, num_tools=0.715, failed=0.27, completion_tokens=43.5]
validation: 80%|████████ | 320/400 [00:18<00:01, 47.62it/s, reward=-0.682, num_turns=1.72, num_tools=0.716, failed=0.269, completion_tokens=43.4]
validation: 80%|████████ | 321/400 [00:18<00:01, 47.62it/s, reward=-0.686, num_turns=1.72, num_tools=0.717, failed=0.268, completion_tokens=43.3]
validation: 80%|████████ | 322/400 [00:18<00:01, 47.62it/s, reward=-0.69, num_turns=1.72, num_tools=0.717, failed=0.267, completion_tokens=43.2]
validation: 81%|████████ | 323/400 [00:18<00:01, 47.62it/s, reward=-0.684, num_turns=1.72, num_tools=0.718, failed=0.266, completion_tokens=43.2]
validation: 81%|████████ | 324/400 [00:18<00:01, 47.62it/s, reward=-0.688, num_turns=1.72, num_tools=0.719, failed=0.265, completion_tokens=43.1][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 81%|████████▏ | 325/400 [00:19<00:01, 47.62it/s, reward=-0.68, num_turns=1.72, num_tools=0.72, failed=0.265, completion_tokens=43]
validation: 82%|████████▏ | 326/400 [00:19<00:01, 47.62it/s, reward=-0.676, num_turns=1.72, num_tools=0.721, failed=0.264, completion_tokens=43.1]
validation: 82%|████████▏ | 327/400 [00:19<00:01, 47.62it/s, reward=-0.68, num_turns=1.72, num_tools=0.722, failed=0.263, completion_tokens=43]
validation: 82%|████████▏ | 328/400 [00:19<00:01, 47.62it/s, reward=-0.666, num_turns=1.72, num_tools=0.723, failed=0.262, completion_tokens=43]
validation: 82%|████████▏ | 329/400 [00:19<00:01, 47.62it/s, reward=-0.673, num_turns=1.72, num_tools=0.72, failed=0.264, completion_tokens=43]
validation: 82%|████████▎ | 330/400 [00:19<00:01, 47.62it/s, reward=-0.668, num_turns=1.72, num_tools=0.721, failed=0.264, completion_tokens=42.9]
validation: 83%|████████▎ | 331/400 [00:19<00:01, 47.62it/s, reward=-0.672, num_turns=1.72, num_tools=0.722, failed=0.263, completion_tokens=42.9]
validation: 83%|████████▎ | 332/400 [00:19<00:01, 47.62it/s, reward=-0.676, num_turns=1.72, num_tools=0.723, failed=0.262, completion_tokens=42.8]
validation: 83%|████████▎ | 333/400 [00:19<00:01, 47.62it/s, reward=-0.68, num_turns=1.72, num_tools=0.724, failed=0.261, completion_tokens=42.8]
validation: 84%|████████▎ | 334/400 [00:19<00:01, 47.62it/s, reward=-0.673, num_turns=1.72, num_tools=0.725, failed=0.26, completion_tokens=42.8]
validation: 84%|████████▍ | 335/400 [00:19<00:01, 47.62it/s, reward=-0.677, num_turns=1.73, num_tools=0.725, failed=0.26, completion_tokens=42.9]
validation: 84%|████████▍ | 336/400 [00:19<00:01, 47.62it/s, reward=-0.673, num_turns=1.73, num_tools=0.726, failed=0.259, completion_tokens=42.9]
validation: 84%|████████▍ | 337/400 [00:19<00:01, 47.62it/s, reward=-0.665, num_turns=1.73, num_tools=0.727, failed=0.258, completion_tokens=42.8]
validation: 84%|████████▍ | 338/400 [00:19<00:01, 47.62it/s, reward=-0.657, num_turns=1.73, num_tools=0.728, failed=0.257, completion_tokens=42.7]
validation: 85%|████████▍ | 339/400 [00:19<00:01, 47.62it/s, reward=-0.661, num_turns=1.73, num_tools=0.729, failed=0.257, completion_tokens=42.7][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
[rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 85%|████████▌ | 340/400 [00:19<00:01, 47.62it/s, reward=-0.668, num_turns=1.73, num_tools=0.729, failed=0.259, completion_tokens=42.6]
validation: 85%|████████▌ | 341/400 [00:19<00:01, 47.62it/s, reward=-0.675, num_turns=1.73, num_tools=0.727, failed=0.261, completion_tokens=42.6]
validation: 86%|████████▌ | 342/400 [00:19<00:01, 47.62it/s, reward=-0.682, num_turns=1.73, num_tools=0.728, failed=0.263, completion_tokens=42.6]
validation: 86%|████████▌ | 343/400 [00:19<00:01, 47.62it/s, reward=-0.689, num_turns=1.73, num_tools=0.729, failed=0.265, completion_tokens=42.5]
validation: 86%|████████▌ | 344/400 [00:19<00:01, 47.62it/s, reward=-0.695, num_turns=1.73, num_tools=0.73, failed=0.267, completion_tokens=42.5]
validation: 86%|████████▋ | 345/400 [00:19<00:01, 47.62it/s, reward=-0.702, num_turns=1.73, num_tools=0.73, failed=0.27, completion_tokens=42.5]
validation: 86%|████████▋ | 346/400 [00:19<00:01, 47.62it/s, reward=-0.688, num_turns=1.73, num_tools=0.731, failed=0.269, completion_tokens=42.4]
validation: 87%|████████▋ | 347/400 [00:19<00:01, 47.62it/s, reward=-0.692, num_turns=1.73, num_tools=0.732, failed=0.268, completion_tokens=42.4]
validation: 87%|████████▋ | 348/400 [00:19<00:01, 47.62it/s, reward=-0.679, num_turns=1.73, num_tools=0.733, failed=0.267, completion_tokens=42.4]
validation: 87%|████████▋ | 349/400 [00:19<00:01, 47.62it/s, reward=-0.685, num_turns=1.73, num_tools=0.734, failed=0.269, completion_tokens=42.4]
validation: 88%|████████▊ | 350/400 [00:19<00:01, 47.62it/s, reward=-0.68, num_turns=1.73, num_tools=0.734, failed=0.269, completion_tokens=42.3]
validation: 88%|████████▊ | 351/400 [00:19<00:01, 47.62it/s, reward=-0.684, num_turns=1.74, num_tools=0.735, failed=0.268, completion_tokens=42.4][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 88%|████████▊ | 352/400 [00:19<00:01, 47.62it/s, reward=-0.691, num_turns=1.73, num_tools=0.733, failed=0.27, completion_tokens=42.4] [rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 88%|████████▊ | 353/400 [00:19<00:00, 47.62it/s, reward=-0.697, num_turns=1.73, num_tools=0.731, failed=0.272, completion_tokens=42.4][rollout] caught BadRequestError: Error code: 400 - {'error': {'message': 'Already borrowed', 'type': 'BadRequestError', 'param': None, 'code': 400}}
validation: 88%|████████▊ | 354/400 [00:19<00:00, 47.62it/s, reward=-0.704, num_turns=1.73, num_tools=0.732, failed=0.274, completion_tokens=42.3]
validation: 89%|████████▉ | 355/400 [00:19<00:00, 75.91it/s, reward=-0.704, num_turns=1.73, num_tools=0.732, failed=0.274, completion_tokens=42.3]
validation: 89%|████████▉ | 355/400 [00:19<00:00, 75.91it/s, reward=-0.699, num_turns=1.73, num_tools=0.732, failed=0.273, completion_tokens=42.3]
validation: 89%|████████▉ | 356/400 [00:19<00:00, 75.91it/s, reward=-0.703, num_turns=1.73, num_tools=0.733, failed=0.272, completion_tokens=42.3]
validation: 89%|████████▉ | 357/400 [00:19<00:00, 75.91it/s, reward=-0.706, num_turns=1.73, num_tools=0.734, failed=0.272, completion_tokens=42.4]
validation: 90%|████████▉ | 358/400 [00:19<00:00, 75.91it/s, reward=-0.697, num_turns=1.73, num_tools=0.735, failed=0.271, completion_tokens=42.3]
validation: 90%|████████▉ | 359/400 [00:19<00:00, 75.91it/s, reward=-0.701, num_turns=1.74, num_tools=0.735, failed=0.27, completion_tokens=42.2]
validation: 90%|█████████ | 360/400 [00:19<00:00, 75.91it/s, reward=-0.694, num_turns=1.74, num_tools=0.736, failed=0.269, completion_tokens=42.1]
validation: 90%|█████████ | 361/400 [00:19<00:00, 75.91it/s, reward=-0.698, num_turns=1.74, num_tools=0.737, failed=0.269, completion_tokens=42.1]
validation: 90%|█████████ | 362/400 [00:19<00:00, 75.91it/s, reward=-0.693, num_turns=1.74, num_tools=0.738, failed=0.268, completion_tokens=42]
validation: 91%|█████████ | 363/400 [00:19<00:00, 75.91it/s, reward=-0.68, num_turns=1.74, num_tools=0.738, failed=0.267, completion_tokens=41.9]
validation: 91%|█████████ | 364/400 [00:19<00:00, 75.91it/s, reward=-0.673, num_turns=1.74, num_tools=0.739, failed=0.266, completion_tokens=41.9]
validation: 91%|█████████▏| 365/400 [00:19<00:00, 75.91it/s, reward=-0.666, num_turns=1.74, num_tools=0.74, failed=0.266, completion_tokens=41.8]
validation: 92%|█████████▏| 366/400 [00:19<00:00, 75.91it/s, reward=-0.658, num_turns=1.74, num_tools=0.74, failed=0.265, completion_tokens=41.7]
validation: 92%|█████████▏| 367/400 [00:19<00:00, 75.91it/s, reward=-0.651, num_turns=1.74, num_tools=0.741, failed=0.264, completion_tokens=41.6]
validation: 92%|█████████▏| 368/400 [00:19<00:00, 75.91it/s, reward=-0.644, num_turns=1.74, num_tools=0.742, failed=0.264, completion_tokens=41.6]
validation: 92%|█████████▏| 369/400 [00:19<00:00, 75.91it/s, reward=-0.64, num_turns=1.74, num_tools=0.743, failed=0.263, completion_tokens=41.5]
validation: 92%|█████████▎| 370/400 [00:19<00:00, 85.29it/s, reward=-0.64, num_turns=1.74, num_tools=0.743, failed=0.263, completion_tokens=41.5]
validation: 92%|█████████▎| 370/400 [00:19<00:00, 85.29it/s, reward=-0.635, num_turns=1.74, num_tools=0.743, failed=0.262, completion_tokens=41.5]
validation: 93%|█████████▎| 371/400 [00:19<00:00, 85.29it/s, reward=-0.639, num_turns=1.74, num_tools=0.744, failed=0.261, completion_tokens=41.4]
validation: 93%|█████████▎| 372/400 [00:19<00:00, 85.29it/s, reward=-0.626, num_turns=1.74, num_tools=0.745, failed=0.261, completion_tokens=41.4]
validation: 93%|█████████▎| 373/400 [00:19<00:00, 85.29it/s, reward=-0.614, num_turns=1.75, num_tools=0.745, failed=0.26, completion_tokens=41.3]
validation: 94%|█████████▎| 374/400 [00:19<00:00, 85.29it/s, reward=-0.607, num_turns=1.75, num_tools=0.746, failed=0.259, completion_tokens=41.3]
validation: 94%|█████████▍| 375/400 [00:19<00:00, 85.29it/s, reward=-0.611, num_turns=1.75, num_tools=0.747, failed=0.259, completion_tokens=41.3]
validation: 94%|█████████▍| 376/400 [00:19<00:00, 85.29it/s, reward=-0.604, num_turns=1.75, num_tools=0.747, failed=0.258, completion_tokens=41.3]
validation: 94%|█████████▍| 377/400 [00:19<00:00, 85.29it/s, reward=-0.607, num_turns=1.75, num_tools=0.748, failed=0.257, completion_tokens=41.3]
validation: 94%|█████████▍| 378/400 [00:19<00:00, 85.29it/s, reward=-0.611, num_turns=1.75, num_tools=0.749, failed=0.257, completion_tokens=41.3]
validation: 95%|█████████▍| 379/400 [00:19<00:00, 85.29it/s, reward=-0.615, num_turns=1.75, num_tools=0.749, failed=0.256, completion_tokens=41.3]
validation: 95%|█████████▌| 380/400 [00:19<00:00, 85.29it/s, reward=-0.618, num_turns=1.75, num_tools=0.75, failed=0.255, completion_tokens=41.3]
validation: 95%|█████████▌| 381/400 [00:19<00:00, 85.29it/s, reward=-0.622, num_turns=1.75, num_tools=0.751, failed=0.255, completion_tokens=41.5]
validation: 96%|█████████▌| 382/400 [00:19<00:00, 85.29it/s, reward=-0.626, num_turns=1.75, num_tools=0.754, failed=0.254, completion_tokens=41.5]
validation: 96%|█████████▌| 383/400 [00:19<00:00, 85.29it/s, reward=-0.62, num_turns=1.75, num_tools=0.755, failed=0.253, completion_tokens=41.6]
validation: 96%|█████████▌| 384/400 [00:19<00:00, 81.11it/s, reward=-0.62, num_turns=1.75, num_tools=0.755, failed=0.253, completion_tokens=41.6]
validation: 96%|█████████▌| 384/400 [00:19<00:00, 81.11it/s, reward=-0.613, num_turns=1.76, num_tools=0.755, failed=0.253, completion_tokens=41.5]
validation: 96%|█████████▋| 385/400 [00:19<00:00, 81.11it/s, reward=-0.617, num_turns=1.76, num_tools=0.756, failed=0.252, completion_tokens=41.5]
validation: 96%|█████████▋| 386/400 [00:19<00:00, 81.11it/s, reward=-0.62, num_turns=1.76, num_tools=0.756, failed=0.251, completion_tokens=41.5]
validation: 97%|█████████▋| 387/400 [00:19<00:00, 81.11it/s, reward=-0.615, num_turns=1.76, num_tools=0.757, failed=0.251, completion_tokens=41.5]
validation: 97%|█████████▋| 388/400 [00:19<00:00, 81.11it/s, reward=-0.619, num_turns=1.76, num_tools=0.758, failed=0.25, completion_tokens=41.4]
validation: 97%|█████████▋| 389/400 [00:19<00:00, 81.11it/s, reward=-0.623, num_turns=1.76, num_tools=0.758, failed=0.249, completion_tokens=41.4]
validation: 98%|█████████▊| 390/400 [00:19<00:00, 81.11it/s, reward=-0.626, num_turns=1.76, num_tools=0.759, failed=0.249, completion_tokens=41.5]
validation: 98%|█████████▊| 391/400 [00:19<00:00, 81.11it/s, reward=-0.623, num_turns=1.76, num_tools=0.76, failed=0.248, completion_tokens=41.6]
validation: 98%|█████████▊| 392/400 [00:19<00:00, 81.11it/s, reward=-0.619, num_turns=1.76, num_tools=0.76, failed=0.247, completion_tokens=41.7]
validation: 98%|█████████▊| 393/400 [00:19<00:00, 81.11it/s, reward=-0.623, num_turns=1.76, num_tools=0.763, failed=0.247, completion_tokens=41.6]
validation: 98%|█████████▊| 394/400 [00:19<00:00, 81.11it/s, reward=-0.616, num_turns=1.76, num_tools=0.764, failed=0.246, completion_tokens=41.7]
validation: 99%|█████████▉| 395/400 [00:19<00:00, 81.11it/s, reward=-0.622, num_turns=1.76, num_tools=0.762, failed=0.246, completion_tokens=42.1]
validation: 99%|█████████▉| 396/400 [00:19<00:00, 61.24it/s, reward=-0.622, num_turns=1.76, num_tools=0.762, failed=0.246, completion_tokens=42.1]
validation: 99%|█████████▉| 396/400 [00:19<00:00, 61.24it/s, reward=-0.618, num_turns=1.76, num_tools=0.763, failed=0.245, completion_tokens=42.1]
validation: 99%|█████████▉| 397/400 [00:19<00:00, 61.24it/s, reward=-0.622, num_turns=1.76, num_tools=0.763, failed=0.244, completion_tokens=42.4]
validation: 100%|█████████▉| 398/400 [00:20<00:00, 61.24it/s, reward=-0.625, num_turns=1.76, num_tools=0.764, failed=0.244, completion_tokens=43.2]
validation: 100%|█████████▉| 399/400 [00:20<00:00, 61.24it/s, reward=-0.629, num_turns=1.76, num_tools=0.764, failed=0.243, completion_tokens=43.6]
validation: 100%|██████████| 400/400 [00:20<00:00, 61.24it/s, reward=-0.632, num_turns=1.76, num_tools=0.765, failed=0.242, completion_tokens=44.1]
validation: 100%|██████████| 400/400 [00:20<00:00, 19.37it/s, reward=-0.632, num_turns=1.76, num_tools=0.765, failed=0.242, completion_tokens=44.1]
Val avg reward: -0.632
============================================================
Step 16/50
============================================================
step 16: 0%| | 0/32 [00:00<?, ?it/s]
step 16: 3%|▎ | 1/32 [00:01<00:31, 1.00s/it]
step 16: 3%|▎ | 1/32 [00:01<00:31, 1.00s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=21]
step 16: 6%|▋ | 2/32 [00:01<00:30, 1.00s/it, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=30]
step 16: 9%|▉ | 3/32 [00:01<00:09, 3.15it/s, reward=-3, num_turns=1, num_tools=0, failed=0, completion_tokens=30]
step 16: 9%|▉ | 3/32 [00:01<00:09, 3.15it/s, reward=-0.667, num_turns=1.33, num_tools=0.333, failed=0, completion_tokens=26.3]
step 16: 12%|█▎ | 4/32 [00:01<00:08, 3.15it/s, reward=-1, num_turns=1.5, num_tools=0.5, failed=0, completion_tokens=25.8]
step 16: 16%|█▌ | 5/32 [00:01<00:08, 3.15it/s, reward=0, num_turns=1.6, num_tools=0.6, failed=0, completion_tokens=25]
step 16: 19%|█▉ | 6/32 [00:01<00:08, 3.15it/s, reward=0.667, num_turns=1.67, num_tools=0.667, failed=0, completion_tokens=24.5]
step 16: 22%|██▏ | 7/32 [00:01<00:07, 3.15it/s, reward=1.14, num_turns=1.71, num_tools=0.714, failed=0, completion_tokens=24.1]
step 16: 25%|██▌ | 8/32 [00:01<00:07, 3.15it/s, reward=1.5, num_turns=1.75, num_tools=0.75, failed=0, completion_tokens=23.9]
step 16: 28%|██▊ | 9/32 [00:01<00:07, 3.15it/s, reward=1.78, num_turns=1.78, num_tools=0.778, failed=0, completion_tokens=24.3]
step 16: 31%|███▏ | 10/32 [00:01<00:06, 3.15it/s, reward=1.7, num_turns=1.8, num_tools=0.8, failed=0, completion_tokens=24.8]
step 16: 34%|███▍ | 11/32 [00:01<00:06, 3.15it/s, reward=1.91, num_turns=1.82, num_tools=0.818, failed=0, completion_tokens=25]
step 16: 38%|███▊ | 12/32 [00:01<00:06, 3.15it/s, reward=2.08, num_turns=1.83, num_tools=0.833, failed=0, completion_tokens=24.9]
step 16: 41%|████ | 13/32 [00:01<00:06, 3.15it/s, reward=2.23, num_turns=1.85, num_tools=0.846, failed=0, completion_tokens=25.1]
step 16: 44%|████▍ | 14/32 [00:01<00:05, 3.15it/s, reward=2.36, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=25.2]
step 16: 47%|████▋ | 15/32 [00:01<00:05, 3.15it/s, reward=2.47, num_turns=1.87, num_tools=0.867, failed=0, completion_tokens=25.4]
step 16: 50%|█████ | 16/32 [00:01<00:05, 3.15it/s, reward=2.56, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=25.6]
step 16: 53%|█████▎ | 17/32 [00:01<00:04, 3.15it/s, reward=2.65, num_turns=1.88, num_tools=0.882, failed=0, completion_tokens=25.3]
step 16: 56%|█████▋ | 18/32 [00:01<00:04, 3.15it/s, reward=2.72, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=25.1]
step 16: 59%|█████▉ | 19/32 [00:01<00:00, 24.85it/s, reward=2.72, num_turns=1.89, num_tools=0.889, failed=0, completion_tokens=25.1]
step 16: 59%|█████▉ | 19/32 [00:01<00:00, 24.85it/s, reward=2.79, num_turns=1.89, num_tools=0.895, failed=0, completion_tokens=25.2]
step 16: 62%|██████▎ | 20/32 [00:01<00:00, 24.85it/s, reward=2.85, num_turns=1.9, num_tools=0.9, failed=0, completion_tokens=25.4]
step 16: 66%|██████▌ | 21/32 [00:01<00:00, 24.85it/s, reward=2.75, num_turns=1.9, num_tools=0.905, failed=0, completion_tokens=26]
step 16: 69%|██████▉ | 22/32 [00:01<00:00, 24.85it/s, reward=2.48, num_turns=1.86, num_tools=0.864, failed=0, completion_tokens=28.9]
step 16: 72%|███████▏ | 23/32 [00:01<00:00, 24.85it/s, reward=2.29, num_turns=1.87, num_tools=0.87, failed=0, completion_tokens=29.2]
step 16: 75%|███████▌ | 24/32 [00:01<00:00, 24.85it/s, reward=2.11, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=29.4]
step 16: 78%|███████▊ | 25/32 [00:01<00:00, 24.85it/s, reward=1.91, num_turns=1.84, num_tools=0.84, failed=0, completion_tokens=32.6]
step 16: 81%|████████▏ | 26/32 [00:01<00:00, 30.29it/s, reward=1.91, num_turns=1.84, num_tools=0.84, failed=0, completion_tokens=32.6]
step 16: 81%|████████▏ | 26/32 [00:01<00:00, 30.29it/s, reward=1.88, num_turns=1.85, num_tools=0.846, failed=0, completion_tokens=32.9]
step 16: 84%|████████▍ | 27/32 [00:01<00:00, 30.29it/s, reward=1.86, num_turns=1.85, num_tools=0.852, failed=0, completion_tokens=33.4]
step 16: 88%|████████▊ | 28/32 [00:01<00:00, 30.29it/s, reward=1.82, num_turns=1.86, num_tools=0.857, failed=0, completion_tokens=34.2]
step 16: 91%|█████████ | 29/32 [00:01<00:00, 30.29it/s, reward=1.79, num_turns=1.86, num_tools=0.862, failed=0, completion_tokens=35.1]
step 16: 94%|█████████▍| 30/32 [00:01<00:00, 30.29it/s, reward=1.67, num_turns=1.87, num_tools=0.867, failed=0, completion_tokens=35.1]
step 16: 97%|█████████▋| 31/32 [00:01<00:00, 30.29it/s, reward=1.65, num_turns=1.87, num_tools=0.871, failed=0, completion_tokens=36.4]
step 16: 100%|██████████| 32/32 [00:01<00:00, 26.13it/s, reward=1.65, num_turns=1.87, num_tools=0.871, failed=0, completion_tokens=36.4]
step 16: 100%|██████████| 32/32 [00:01<00:00, 26.13it/s, reward=1.61, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=38]
step 16: 100%|██████████| 32/32 [00:01<00:00, 18.60it/s, reward=1.61, num_turns=1.88, num_tools=0.875, failed=0, completion_tokens=38]
group 0: mean=+4.00 std=0.000 min=+4.0 max=+4.0 | What's the weather like in London?
group 1: mean=-1.12 std=1.699 min=-3.0 max=+1.3 | What is Japan's population density in people per s
group 2: mean=+4.00 std=0.000 min=+4.0 max=+4.0 | Convert 23 kg to lbs.
group 3: mean=-0.42 std=1.778 min=-3.0 max=+1.3 | What is India's population density in people per s
Avg reward: 1.615 | Avg tools/rollout: 0.9 | groups with variance: 2/4
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0014
Deleted checkpoint ./.art/rl-tool-use/models/qwen-0.5b-tool-agent/checkpoints/0010
Packed 16 trajectories into 2 sequences of length 2048
train: 0%| | 0/2 [00:00<?, ?it/s][2026-04-13 02:45:31] INFO _base_client.py:1693: Retrying request to /completions in 0.489393 seconds
[2026-04-13 02:45:36] INFO _base_client.py:1693: Retrying request to /completions in 0.916879 seconds
[2026-04-13 02:46:17] INFO _base_client.py:1693: Retrying request to /completions in 0.386873 seconds
[2026-04-13 02:46:23] INFO _base_client.py:1693: Retrying request to /completions in 0.894120 seconds
[train.py] suppressed _monitor_openai_server crash: APITimeoutError: Request timed out.
Traceback (most recent call last):
File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "/workspace/RL-Trained-Tool-Use-Agent/src/train.py", line 211, in <module>
main()
File "/workspace/RL-Trained-Tool-Use-Agent/src/train.py", line 207, in main
asyncio.run(train(**kwargs))
File "/usr/local/lib/python3.12/dist-packages/nest_asyncio.py", line 30, in run
return loop.run_until_complete(task)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/nest_asyncio.py", line 98, in run_until_complete
return f.result()
^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/futures.py", line 203, in result
raise self._exception.with_traceback(self._exception_tb)
File "/usr/lib/python3.12/asyncio/tasks.py", line 316, in __step_run_and_handle_result
result = coro.throw(exc)
^^^^^^^^^^^^^^^
File "/workspace/RL-Trained-Tool-Use-Agent/src/train.py", line 153, in train
result = await backend.train(model, train_groups, learning_rate=learning_rate)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/art/local/backend.py", line 644, in train
async for metrics in self._train_model(
File "/usr/local/lib/python3.12/dist-packages/art/local/backend.py", line 783, in _train_model
async for result in service.train(
File "/usr/local/lib/python3.12/dist-packages/mp_actors/move.py", line 226, in async_gen_wrapper
send_value = yield await asyncio.wrap_future(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/futures.py", line 287, in __await__
yield self # This tells Task to wait for completion.
^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/tasks.py", line 385, in __wakeup
future.result()
File "/usr/lib/python3.12/asyncio/futures.py", line 203, in result
raise self._exception.with_traceback(self._exception_tb)
RuntimeError: Proxy is closing
train: 0%| | 0/2 [01:06<?, ?it/s]
Reference in New Issue View Git Blame Copy Permalink