echo / MVP /distill.log
void0x14
feat: multimodal model anahtar teslim + test + rapor
598b018 unverified
Raw History Blame Contribute Delete
157 kB
The fast path is not available because one of the required library is not installed. Falling back to torch implementation. To install follow https://github.com/fla-org/flash-linear-attention#installation and https://github.com/Dao-AILab/causal-conv1d
Loading teacher from /home/void0x14/.cache/huggingface/hub/models--Qwen--Qwen3.5-0.8B-Base/snapshots/dc7cdfe2ee4154fa7e30f5b51ca41bfa40174e68...
Loading weights: 0%| | 0/473 [00:00<?, ?it/s] Loading weights: 0%| | 1/473 [00:00<00:00, 7371.36it/s, Materializing param=language_model.embed_tokens.weight] Loading weights: 0%| | 1/473 [00:00<00:00, 3130.08it/s, Materializing param=language_model.embed_tokens.weight] Loading weights: 0%| | 2/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.embed_tokens.weight] Loading weights: 0%| | 2/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.layers.0.input_layernorm.weight] Loading weights: 0%| | 2/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.layers.0.input_layernorm.weight] Loading weights: 1%| | 3/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.A_log] Loading weights: 1%| | 3/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.A_log] Loading weights: 1%| | 4/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.conv1d.weight] Loading weights: 1%| | 4/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.conv1d.weight] Loading weights: 1%| | 5/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.dt_bias] Loading weights: 1%| | 5/473 [00:00<00:59, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.dt_bias] Loading weights: 1%|▏ | 6/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.in_proj_a.weight] Loading weights: 1%|▏ | 6/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.in_proj_a.weight] Loading weights: 1%|▏ | 7/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.in_proj_b.weight] Loading weights: 1%|▏ | 7/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.in_proj_b.weight] Loading weights: 2%|▏ | 8/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.in_proj_qkv.weight] Loading weights: 2%|▏ | 8/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.in_proj_qkv.weight] Loading weights: 2%|▏ | 9/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.in_proj_z.weight] Loading weights: 2%|▏ | 9/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.in_proj_z.weight] Loading weights: 2%|▏ | 10/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.norm.weight] Loading weights: 2%|▏ | 10/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.norm.weight] Loading weights: 2%|▏ | 11/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.out_proj.weight] Loading weights: 2%|▏ | 11/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.linear_attn.out_proj.weight] Loading weights: 3%|β–Ž | 12/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.mlp.down_proj.weight] Loading weights: 3%|β–Ž | 12/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.mlp.down_proj.weight] Loading weights: 3%|β–Ž | 13/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.mlp.gate_proj.weight] Loading weights: 3%|β–Ž | 13/473 [00:00<00:58, 7.93it/s, Materializing param=language_model.layers.0.mlp.gate_proj.weight] Loading weights: 3%|β–Ž | 14/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.0.mlp.up_proj.weight] Loading weights: 3%|β–Ž | 14/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.0.mlp.up_proj.weight] Loading weights: 3%|β–Ž | 15/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.0.post_attention_layernorm.weight] Loading weights: 3%|β–Ž | 15/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.0.post_attention_layernorm.weight] Loading weights: 3%|β–Ž | 16/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.input_layernorm.weight] Loading weights: 3%|β–Ž | 16/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.input_layernorm.weight] Loading weights: 4%|β–Ž | 17/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.A_log] Loading weights: 4%|β–Ž | 17/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.A_log] Loading weights: 4%|▍ | 18/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.conv1d.weight] Loading weights: 4%|▍ | 18/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.conv1d.weight] Loading weights: 4%|▍ | 19/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.dt_bias] Loading weights: 4%|▍ | 19/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.dt_bias] Loading weights: 4%|▍ | 20/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.in_proj_a.weight] Loading weights: 4%|▍ | 20/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.in_proj_a.weight] Loading weights: 4%|▍ | 21/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.in_proj_b.weight] Loading weights: 4%|▍ | 21/473 [00:00<00:57, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.in_proj_b.weight] Loading weights: 5%|▍ | 22/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.in_proj_qkv.weight] Loading weights: 5%|▍ | 22/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.in_proj_qkv.weight] Loading weights: 5%|▍ | 23/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.in_proj_z.weight] Loading weights: 5%|▍ | 23/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.in_proj_z.weight] Loading weights: 5%|β–Œ | 24/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.norm.weight] Loading weights: 5%|β–Œ | 24/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.norm.weight] Loading weights: 5%|β–Œ | 25/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.out_proj.weight] Loading weights: 5%|β–Œ | 25/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.linear_attn.out_proj.weight] Loading weights: 5%|β–Œ | 26/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.mlp.down_proj.weight] Loading weights: 5%|β–Œ | 26/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.mlp.down_proj.weight] Loading weights: 6%|β–Œ | 27/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.mlp.gate_proj.weight] Loading weights: 6%|β–Œ | 27/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.mlp.gate_proj.weight] Loading weights: 6%|β–Œ | 28/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.mlp.up_proj.weight] Loading weights: 6%|β–Œ | 28/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.mlp.up_proj.weight] Loading weights: 6%|β–Œ | 29/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.post_attention_layernorm.weight] Loading weights: 6%|β–Œ | 29/473 [00:00<00:56, 7.93it/s, Materializing param=language_model.layers.1.post_attention_layernorm.weight] Loading weights: 6%|β–‹ | 30/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.input_layernorm.weight] Loading weights: 6%|β–‹ | 30/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.input_layernorm.weight] Loading weights: 7%|β–‹ | 31/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.A_log] Loading weights: 7%|β–‹ | 31/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.A_log] Loading weights: 7%|β–‹ | 32/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.conv1d.weight] Loading weights: 7%|β–‹ | 32/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.conv1d.weight] Loading weights: 7%|β–‹ | 33/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.dt_bias] Loading weights: 7%|β–‹ | 33/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.dt_bias] Loading weights: 7%|β–‹ | 34/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.in_proj_a.weight] Loading weights: 7%|β–‹ | 34/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.in_proj_a.weight] Loading weights: 7%|β–‹ | 35/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.in_proj_b.weight] Loading weights: 7%|β–‹ | 35/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.in_proj_b.weight] Loading weights: 8%|β–Š | 36/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.in_proj_qkv.weight] Loading weights: 8%|β–Š | 36/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.in_proj_qkv.weight] Loading weights: 8%|β–Š | 37/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.in_proj_z.weight] Loading weights: 8%|β–Š | 37/473 [00:00<00:55, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.in_proj_z.weight] Loading weights: 8%|β–Š | 38/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.norm.weight] Loading weights: 8%|β–Š | 38/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.norm.weight] Loading weights: 8%|β–Š | 39/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.out_proj.weight] Loading weights: 8%|β–Š | 39/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.linear_attn.out_proj.weight] Loading weights: 8%|β–Š | 40/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.mlp.down_proj.weight] Loading weights: 8%|β–Š | 40/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.mlp.down_proj.weight] Loading weights: 9%|β–Š | 41/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.mlp.gate_proj.weight] Loading weights: 9%|β–Š | 41/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.mlp.gate_proj.weight] Loading weights: 9%|β–‰ | 42/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.mlp.up_proj.weight] Loading weights: 9%|β–‰ | 42/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.mlp.up_proj.weight] Loading weights: 9%|β–‰ | 43/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.post_attention_layernorm.weight] Loading weights: 9%|β–‰ | 43/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.2.post_attention_layernorm.weight] Loading weights: 9%|β–‰ | 44/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.3.input_layernorm.weight] Loading weights: 9%|β–‰ | 44/473 [00:00<00:54, 7.93it/s, Materializing param=language_model.layers.3.input_layernorm.weight] Loading weights: 10%|β–‰ | 45/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.mlp.down_proj.weight] Loading weights: 10%|β–‰ | 45/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.mlp.down_proj.weight] Loading weights: 10%|β–‰ | 46/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.mlp.gate_proj.weight] Loading weights: 10%|β–‰ | 46/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.mlp.gate_proj.weight] Loading weights: 10%|β–‰ | 47/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.mlp.up_proj.weight] Loading weights: 10%|β–‰ | 47/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.mlp.up_proj.weight] Loading weights: 10%|β–ˆ | 48/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.post_attention_layernorm.weight] Loading weights: 10%|β–ˆ | 48/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.post_attention_layernorm.weight] Loading weights: 10%|β–ˆ | 49/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.self_attn.k_norm.weight] Loading weights: 10%|β–ˆ | 49/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.self_attn.k_norm.weight] Loading weights: 11%|β–ˆ | 50/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.self_attn.k_proj.weight] Loading weights: 11%|β–ˆ | 50/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.self_attn.k_proj.weight] Loading weights: 11%|β–ˆ | 51/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.self_attn.o_proj.weight] Loading weights: 11%|β–ˆ | 51/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.self_attn.o_proj.weight] Loading weights: 11%|β–ˆ | 52/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.self_attn.q_norm.weight] Loading weights: 11%|β–ˆ | 52/473 [00:00<00:53, 7.93it/s, Materializing param=language_model.layers.3.self_attn.q_norm.weight] Loading weights: 11%|β–ˆ | 53/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.3.self_attn.q_proj.weight] Loading weights: 11%|β–ˆ | 53/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.3.self_attn.q_proj.weight] Loading weights: 11%|β–ˆβ– | 54/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.3.self_attn.v_proj.weight] Loading weights: 11%|β–ˆβ– | 54/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.3.self_attn.v_proj.weight] Loading weights: 12%|β–ˆβ– | 55/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.input_layernorm.weight] Loading weights: 12%|β–ˆβ– | 55/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.input_layernorm.weight] Loading weights: 12%|β–ˆβ– | 56/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.A_log] Loading weights: 12%|β–ˆβ– | 56/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.A_log] Loading weights: 12%|β–ˆβ– | 57/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.conv1d.weight] Loading weights: 12%|β–ˆβ– | 57/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.conv1d.weight] Loading weights: 12%|β–ˆβ– | 58/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.dt_bias] Loading weights: 12%|β–ˆβ– | 58/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.dt_bias] Loading weights: 12%|β–ˆβ– | 59/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.in_proj_a.weight] Loading weights: 12%|β–ˆβ– | 59/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.in_proj_a.weight] Loading weights: 13%|β–ˆβ–Ž | 60/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.in_proj_b.weight] Loading weights: 13%|β–ˆβ–Ž | 60/473 [00:00<00:52, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.in_proj_b.weight] Loading weights: 13%|β–ˆβ–Ž | 61/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.in_proj_qkv.weight] Loading weights: 13%|β–ˆβ–Ž | 61/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.in_proj_qkv.weight] Loading weights: 13%|β–ˆβ–Ž | 62/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.in_proj_z.weight] Loading weights: 13%|β–ˆβ–Ž | 62/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.in_proj_z.weight] Loading weights: 13%|β–ˆβ–Ž | 63/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.norm.weight] Loading weights: 13%|β–ˆβ–Ž | 63/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.norm.weight] Loading weights: 14%|β–ˆβ–Ž | 64/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.out_proj.weight] Loading weights: 14%|β–ˆβ–Ž | 64/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.linear_attn.out_proj.weight] Loading weights: 14%|β–ˆβ–Ž | 65/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.mlp.down_proj.weight] Loading weights: 14%|β–ˆβ–Ž | 65/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.mlp.down_proj.weight] Loading weights: 14%|β–ˆβ– | 66/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.mlp.gate_proj.weight] Loading weights: 14%|β–ˆβ– | 66/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.mlp.gate_proj.weight] Loading weights: 14%|β–ˆβ– | 67/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.mlp.up_proj.weight] Loading weights: 14%|β–ˆβ– | 67/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.mlp.up_proj.weight] Loading weights: 14%|β–ˆβ– | 68/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.post_attention_layernorm.weight] Loading weights: 14%|β–ˆβ– | 68/473 [00:00<00:51, 7.93it/s, Materializing param=language_model.layers.4.post_attention_layernorm.weight] Loading weights: 15%|β–ˆβ– | 69/473 [00:00<00:50, 7.93it/s, Materializing param=language_model.layers.5.input_layernorm.weight] Loading weights: 15%|β–ˆβ– | 69/473 [00:00<00:50, 7.93it/s, Materializing param=language_model.layers.5.input_layernorm.weight] Loading weights: 15%|β–ˆβ– | 70/473 [00:00<00:50, 7.93it/s, Materializing param=language_model.layers.5.linear_attn.A_log] Loading weights: 15%|β–ˆβ– | 70/473 [00:00<00:50, 7.93it/s, Materializing param=language_model.layers.5.linear_attn.A_log] Loading weights: 15%|β–ˆβ–Œ | 71/473 [00:00<00:50, 7.93it/s, Materializing param=language_model.layers.5.linear_attn.conv1d.weight] Loading weights: 15%|β–ˆβ–Œ | 71/473 [00:00<00:50, 7.93it/s, Materializing param=language_model.layers.5.linear_attn.conv1d.weight] Loading weights: 15%|β–ˆβ–Œ | 72/473 [00:00<00:50, 7.93it/s, Materializing param=language_model.layers.5.linear_attn.dt_bias] Loading weights: 15%|β–ˆβ–Œ | 72/473 [00:00<00:50, 7.93it/s, Materializing param=language_model.layers.5.linear_attn.dt_bias] Loading weights: 15%|β–ˆβ–Œ | 73/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.dt_bias] Loading weights: 15%|β–ˆβ–Œ | 73/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.in_proj_a.weight] Loading weights: 15%|β–ˆβ–Œ | 73/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.in_proj_a.weight] Loading weights: 16%|β–ˆβ–Œ | 74/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.in_proj_b.weight] Loading weights: 16%|β–ˆβ–Œ | 74/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.in_proj_b.weight] Loading weights: 16%|β–ˆβ–Œ | 75/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.in_proj_qkv.weight] Loading weights: 16%|β–ˆβ–Œ | 75/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.in_proj_qkv.weight] Loading weights: 16%|β–ˆβ–Œ | 76/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.in_proj_z.weight] Loading weights: 16%|β–ˆβ–Œ | 76/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.in_proj_z.weight] Loading weights: 16%|β–ˆβ–‹ | 77/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.norm.weight] Loading weights: 16%|β–ˆβ–‹ | 77/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.norm.weight] Loading weights: 16%|β–ˆβ–‹ | 78/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.out_proj.weight] Loading weights: 16%|β–ˆβ–‹ | 78/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.linear_attn.out_proj.weight] Loading weights: 17%|β–ˆβ–‹ | 79/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.mlp.down_proj.weight] Loading weights: 17%|β–ˆβ–‹ | 79/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.mlp.down_proj.weight] Loading weights: 17%|β–ˆβ–‹ | 80/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.mlp.gate_proj.weight] Loading weights: 17%|β–ˆβ–‹ | 80/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.mlp.gate_proj.weight] Loading weights: 17%|β–ˆβ–‹ | 81/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.mlp.up_proj.weight] Loading weights: 17%|β–ˆβ–‹ | 81/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.mlp.up_proj.weight] Loading weights: 17%|β–ˆβ–‹ | 82/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.post_attention_layernorm.weight] Loading weights: 17%|β–ˆβ–‹ | 82/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.5.post_attention_layernorm.weight] Loading weights: 18%|β–ˆβ–Š | 83/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.input_layernorm.weight] Loading weights: 18%|β–ˆβ–Š | 83/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.input_layernorm.weight] Loading weights: 18%|β–ˆβ–Š | 84/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.A_log] Loading weights: 18%|β–ˆβ–Š | 84/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.A_log] Loading weights: 18%|β–ˆβ–Š | 85/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.conv1d.weight] Loading weights: 18%|β–ˆβ–Š | 85/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.conv1d.weight] Loading weights: 18%|β–ˆβ–Š | 86/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.dt_bias] Loading weights: 18%|β–ˆβ–Š | 86/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.dt_bias] Loading weights: 18%|β–ˆβ–Š | 87/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.in_proj_a.weight] Loading weights: 18%|β–ˆβ–Š | 87/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.in_proj_a.weight] Loading weights: 19%|β–ˆβ–Š | 88/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.in_proj_b.weight] Loading weights: 19%|β–ˆβ–Š | 88/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.in_proj_b.weight] Loading weights: 19%|β–ˆβ–‰ | 89/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.in_proj_qkv.weight] Loading weights: 19%|β–ˆβ–‰ | 89/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.in_proj_qkv.weight] Loading weights: 19%|β–ˆβ–‰ | 90/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.in_proj_z.weight] Loading weights: 19%|β–ˆβ–‰ | 90/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.in_proj_z.weight] Loading weights: 19%|β–ˆβ–‰ | 91/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.norm.weight] Loading weights: 19%|β–ˆβ–‰ | 91/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.norm.weight] Loading weights: 19%|β–ˆβ–‰ | 92/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.out_proj.weight] Loading weights: 19%|β–ˆβ–‰ | 92/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.linear_attn.out_proj.weight] Loading weights: 20%|β–ˆβ–‰ | 93/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.mlp.down_proj.weight] Loading weights: 20%|β–ˆβ–‰ | 93/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.mlp.down_proj.weight] Loading weights: 20%|β–ˆβ–‰ | 94/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.mlp.gate_proj.weight] Loading weights: 20%|β–ˆβ–‰ | 94/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.mlp.gate_proj.weight] Loading weights: 20%|β–ˆβ–ˆ | 95/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.mlp.up_proj.weight] Loading weights: 20%|β–ˆβ–ˆ | 95/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.mlp.up_proj.weight] Loading weights: 20%|β–ˆβ–ˆ | 96/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.post_attention_layernorm.weight] Loading weights: 20%|β–ˆβ–ˆ | 96/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.6.post_attention_layernorm.weight] Loading weights: 21%|β–ˆβ–ˆ | 97/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.input_layernorm.weight] Loading weights: 21%|β–ˆβ–ˆ | 97/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.input_layernorm.weight] Loading weights: 21%|β–ˆβ–ˆ | 98/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.mlp.down_proj.weight] Loading weights: 21%|β–ˆβ–ˆ | 98/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.mlp.down_proj.weight] Loading weights: 21%|β–ˆβ–ˆ | 99/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.mlp.gate_proj.weight] Loading weights: 21%|β–ˆβ–ˆ | 99/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.mlp.gate_proj.weight] Loading weights: 21%|β–ˆβ–ˆ | 100/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.mlp.up_proj.weight] Loading weights: 21%|β–ˆβ–ˆ | 100/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.mlp.up_proj.weight] Loading weights: 21%|β–ˆβ–ˆβ– | 101/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.post_attention_layernorm.weight] Loading weights: 21%|β–ˆβ–ˆβ– | 101/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.post_attention_layernorm.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 102/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.k_norm.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 102/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.k_norm.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 103/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.k_proj.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 103/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.k_proj.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 104/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.o_proj.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 104/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.o_proj.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 105/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.q_norm.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 105/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.q_norm.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 106/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.q_proj.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 106/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.q_proj.weight] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 107/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.v_proj.weight] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 107/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.7.self_attn.v_proj.weight] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 108/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.input_layernorm.weight] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 108/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.input_layernorm.weight] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 109/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.A_log] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 109/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.A_log] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 110/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.conv1d.weight] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 110/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.conv1d.weight] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 111/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.dt_bias] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 111/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.dt_bias] Loading weights: 24%|β–ˆβ–ˆβ–Ž | 112/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_a.weight] Loading weights: 24%|β–ˆβ–ˆβ–Ž | 112/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_a.weight] Loading weights: 24%|β–ˆβ–ˆβ– | 113/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_b.weight] Loading weights: 24%|β–ˆβ–ˆβ– | 113/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_b.weight] Loading weights: 24%|β–ˆβ–ˆβ– | 114/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_qkv.weight] Loading weights: 24%|β–ˆβ–ˆβ– | 114/473 [00:00<00:01, 260.07it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_qkv.weight] Loading weights: 24%|β–ˆβ–ˆβ– | 115/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_qkv.weight] Loading weights: 24%|β–ˆβ–ˆβ– | 115/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_z.weight] Loading weights: 24%|β–ˆβ–ˆβ– | 115/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.linear_attn.in_proj_z.weight] Loading weights: 25%|β–ˆβ–ˆβ– | 116/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.linear_attn.norm.weight] Loading weights: 25%|β–ˆβ–ˆβ– | 116/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.linear_attn.norm.weight] Loading weights: 25%|β–ˆβ–ˆβ– | 117/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.linear_attn.out_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ– | 117/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.linear_attn.out_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ– | 118/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.mlp.down_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ– | 118/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.mlp.down_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ–Œ | 119/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.mlp.gate_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ–Œ | 119/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.mlp.gate_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ–Œ | 120/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.mlp.up_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ–Œ | 120/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.mlp.up_proj.weight] Loading weights: 26%|β–ˆβ–ˆβ–Œ | 121/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.post_attention_layernorm.weight] Loading weights: 26%|β–ˆβ–ˆβ–Œ | 121/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.8.post_attention_layernorm.weight] Loading weights: 26%|β–ˆβ–ˆβ–Œ | 122/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.input_layernorm.weight] Loading weights: 26%|β–ˆβ–ˆβ–Œ | 122/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.input_layernorm.weight] Loading weights: 26%|β–ˆβ–ˆβ–Œ | 123/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.A_log] Loading weights: 26%|β–ˆβ–ˆβ–Œ | 123/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.A_log] Loading weights: 26%|β–ˆβ–ˆβ–Œ | 124/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.conv1d.weight] Loading weights: 26%|β–ˆβ–ˆβ–Œ | 124/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.conv1d.weight] Loading weights: 26%|β–ˆβ–ˆβ–‹ | 125/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.dt_bias] Loading weights: 26%|β–ˆβ–ˆβ–‹ | 125/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.dt_bias] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 126/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.in_proj_a.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 126/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.in_proj_a.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 127/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.in_proj_b.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 127/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.in_proj_b.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 128/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.in_proj_qkv.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 128/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.in_proj_qkv.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 129/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.in_proj_z.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 129/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.in_proj_z.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 130/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.norm.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 130/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.norm.weight] Loading weights: 28%|β–ˆβ–ˆβ–Š | 131/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.out_proj.weight] Loading weights: 28%|β–ˆβ–ˆβ–Š | 131/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.linear_attn.out_proj.weight] Loading weights: 28%|β–ˆβ–ˆβ–Š | 132/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.mlp.down_proj.weight] Loading weights: 28%|β–ˆβ–ˆβ–Š | 132/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.mlp.down_proj.weight] Loading weights: 28%|β–ˆβ–ˆβ–Š | 133/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.mlp.gate_proj.weight] Loading weights: 28%|β–ˆβ–ˆβ–Š | 133/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.mlp.gate_proj.weight] Loading weights: 28%|β–ˆβ–ˆβ–Š | 134/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.mlp.up_proj.weight] Loading weights: 28%|β–ˆβ–ˆβ–Š | 134/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.mlp.up_proj.weight] Loading weights: 29%|β–ˆβ–ˆβ–Š | 135/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.post_attention_layernorm.weight] Loading weights: 29%|β–ˆβ–ˆβ–Š | 135/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.9.post_attention_layernorm.weight] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 136/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.input_layernorm.weight] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 136/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.input_layernorm.weight] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 137/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.A_log] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 137/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.A_log] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 138/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.conv1d.weight] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 138/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.conv1d.weight] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 139/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.dt_bias] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 139/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.dt_bias] Loading weights: 30%|β–ˆβ–ˆβ–‰ | 140/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.in_proj_a.weight] Loading weights: 30%|β–ˆβ–ˆβ–‰ | 140/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.in_proj_a.weight] Loading weights: 30%|β–ˆβ–ˆβ–‰ | 141/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.in_proj_b.weight] Loading weights: 30%|β–ˆβ–ˆβ–‰ | 141/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.in_proj_b.weight] Loading weights: 30%|β–ˆβ–ˆβ–ˆ | 142/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.in_proj_qkv.weight] Loading weights: 30%|β–ˆβ–ˆβ–ˆ | 142/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.in_proj_qkv.weight] Loading weights: 30%|β–ˆβ–ˆβ–ˆ | 143/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.in_proj_z.weight] Loading weights: 30%|β–ˆβ–ˆβ–ˆ | 143/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.in_proj_z.weight] Loading weights: 30%|β–ˆβ–ˆβ–ˆ | 144/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.norm.weight] Loading weights: 30%|β–ˆβ–ˆβ–ˆ | 144/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.norm.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆ | 145/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.out_proj.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆ | 145/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.linear_attn.out_proj.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆ | 146/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.mlp.down_proj.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆ | 146/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.mlp.down_proj.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆ | 147/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.mlp.gate_proj.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆ | 147/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.mlp.gate_proj.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆβ– | 148/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.mlp.up_proj.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆβ– | 148/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.mlp.up_proj.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 149/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.post_attention_layernorm.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 149/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.10.post_attention_layernorm.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 150/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.input_layernorm.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 150/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.input_layernorm.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 151/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.mlp.down_proj.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 151/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.mlp.down_proj.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 152/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.mlp.gate_proj.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 152/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.mlp.gate_proj.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 153/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.mlp.up_proj.weight] Loading weights: 32%|β–ˆβ–ˆβ–ˆβ– | 153/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.mlp.up_proj.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 154/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.post_attention_layernorm.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 154/473 [00:00<00:01, 287.24it/s, Materializing param=language_model.layers.11.post_attention_layernorm.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 155/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.post_attention_layernorm.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 155/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.k_norm.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 155/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.k_norm.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 156/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.k_proj.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 156/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.k_proj.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 157/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.o_proj.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 157/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.o_proj.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 158/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.q_norm.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 158/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.q_norm.weight] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ–Ž | 159/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.q_proj.weight] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ–Ž | 159/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.q_proj.weight] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ– | 160/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.v_proj.weight] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ– | 160/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.11.self_attn.v_proj.weight] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ– | 161/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.input_layernorm.weight] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ– | 161/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.input_layernorm.weight] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ– | 162/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.A_log] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ– | 162/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.A_log] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ– | 163/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.conv1d.weight] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ– | 163/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.conv1d.weight] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ– | 164/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.dt_bias] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ– | 164/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.dt_bias] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ– | 165/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.in_proj_a.weight] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ– | 165/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.in_proj_a.weight] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 166/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.in_proj_b.weight] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 166/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.in_proj_b.weight] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 167/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.in_proj_qkv.weight] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 167/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.in_proj_qkv.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 168/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.in_proj_z.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 168/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.in_proj_z.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 169/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.norm.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 169/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.norm.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 170/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.out_proj.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 170/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.linear_attn.out_proj.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 171/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.mlp.down_proj.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 171/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.mlp.down_proj.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–‹ | 172/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.mlp.gate_proj.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–‹ | 172/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.mlp.gate_proj.weight] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 173/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.mlp.up_proj.weight] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 173/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.mlp.up_proj.weight] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 174/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.post_attention_layernorm.weight] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 174/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.12.post_attention_layernorm.weight] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 175/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.input_layernorm.weight] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 175/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.input_layernorm.weight] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 176/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.A_log] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 176/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.A_log] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 177/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.conv1d.weight] Loading weights: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 177/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.conv1d.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 178/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.dt_bias] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 178/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.dt_bias] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 179/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.in_proj_a.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 179/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.in_proj_a.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 180/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.in_proj_b.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 180/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.in_proj_b.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 181/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.in_proj_qkv.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 181/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.in_proj_qkv.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 182/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.in_proj_z.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 182/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.in_proj_z.weight] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–Š | 183/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.norm.weight] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–Š | 183/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.norm.weight] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 184/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.out_proj.weight] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 184/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.linear_attn.out_proj.weight] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 185/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.mlp.down_proj.weight] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 185/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.mlp.down_proj.weight] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 186/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.mlp.gate_proj.weight] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 186/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.mlp.gate_proj.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 187/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.mlp.up_proj.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 187/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.mlp.up_proj.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 188/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.post_attention_layernorm.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 188/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.13.post_attention_layernorm.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 189/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.input_layernorm.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 189/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.input_layernorm.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 190/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.A_log] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 190/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.A_log] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 191/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.conv1d.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 191/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.conv1d.weight] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 192/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.dt_bias] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 192/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.dt_bias] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 193/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.in_proj_a.weight] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 193/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.in_proj_a.weight] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 194/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.in_proj_b.weight] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 194/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.in_proj_b.weight] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 195/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.in_proj_qkv.weight] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 195/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.in_proj_qkv.weight] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 196/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.in_proj_z.weight] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 196/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.in_proj_z.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 197/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.norm.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 197/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.norm.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 198/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.out_proj.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 198/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.linear_attn.out_proj.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 199/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.mlp.down_proj.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 199/473 [00:00<00:00, 321.81it/s, Materializing param=language_model.layers.14.mlp.down_proj.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 200/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.14.mlp.down_proj.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 200/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.14.mlp.gate_proj.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 200/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.14.mlp.gate_proj.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 201/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.14.mlp.up_proj.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 201/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.14.mlp.up_proj.weight] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 202/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.14.post_attention_layernorm.weight] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 202/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.14.post_attention_layernorm.weight] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 203/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.input_layernorm.weight] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 203/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.input_layernorm.weight] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 204/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.mlp.down_proj.weight] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 204/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.mlp.down_proj.weight] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 205/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.mlp.gate_proj.weight] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 205/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.mlp.gate_proj.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 206/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.mlp.up_proj.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 206/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.mlp.up_proj.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 207/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.post_attention_layernorm.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 207/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.post_attention_layernorm.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 208/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.k_norm.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 208/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.k_norm.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 209/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.k_proj.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 209/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.k_proj.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 210/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.o_proj.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 210/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.o_proj.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 211/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.q_norm.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 211/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.q_norm.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 212/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.q_proj.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 212/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.q_proj.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 213/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.v_proj.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 213/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.15.self_attn.v_proj.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 214/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.input_layernorm.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 214/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.input_layernorm.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 215/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.A_log] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 215/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.A_log] Loading weights: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 216/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.conv1d.weight] Loading weights: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 216/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.conv1d.weight] Loading weights: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 217/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.dt_bias] Loading weights: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 217/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.dt_bias] Loading weights: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 218/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.in_proj_a.weight] Loading weights: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 218/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.in_proj_a.weight] Loading weights: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 219/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.in_proj_b.weight] Loading weights: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 219/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.in_proj_b.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 220/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.in_proj_qkv.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 220/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.in_proj_qkv.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 221/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.in_proj_z.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 221/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.in_proj_z.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 222/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.norm.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 222/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.norm.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 223/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.out_proj.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 223/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.linear_attn.out_proj.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 224/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.mlp.down_proj.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 224/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.mlp.down_proj.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 225/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.mlp.gate_proj.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 225/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.mlp.gate_proj.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 226/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.mlp.up_proj.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 226/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.mlp.up_proj.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 227/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.post_attention_layernorm.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 227/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.16.post_attention_layernorm.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 228/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.input_layernorm.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 228/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.input_layernorm.weight] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 229/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.A_log] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 229/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.A_log] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 230/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.conv1d.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 230/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.conv1d.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 231/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.dt_bias] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 231/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.dt_bias] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 232/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.in_proj_a.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 232/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.in_proj_a.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 233/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.in_proj_b.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 233/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.in_proj_b.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 234/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.in_proj_qkv.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 234/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.in_proj_qkv.weight] Loading weights: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 235/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.in_proj_z.weight] Loading weights: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 235/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.in_proj_z.weight] Loading weights: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 236/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.norm.weight] Loading weights: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 236/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.norm.weight] Loading weights: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 237/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.out_proj.weight] Loading weights: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 237/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.linear_attn.out_proj.weight] Loading weights: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 238/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.mlp.down_proj.weight] Loading weights: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 238/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.mlp.down_proj.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 239/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.mlp.gate_proj.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 239/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.mlp.gate_proj.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 240/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.mlp.up_proj.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 240/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.mlp.up_proj.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 241/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.post_attention_layernorm.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 241/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.17.post_attention_layernorm.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 242/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.input_layernorm.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 242/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.input_layernorm.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 243/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.A_log] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 243/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.A_log] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 244/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.conv1d.weight] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 244/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.conv1d.weight] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 245/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.dt_bias] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 245/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.dt_bias] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 246/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.in_proj_a.weight] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 246/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.in_proj_a.weight] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 247/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.in_proj_b.weight] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 247/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.in_proj_b.weight] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 248/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.in_proj_qkv.weight] Loading weights: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 248/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.in_proj_qkv.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 249/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.in_proj_z.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 249/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.in_proj_z.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 250/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.norm.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 250/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.norm.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 251/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.out_proj.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 251/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.linear_attn.out_proj.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 252/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.mlp.down_proj.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 252/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.mlp.down_proj.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 253/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.mlp.gate_proj.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 253/473 [00:00<00:00, 359.68it/s, Materializing param=language_model.layers.18.mlp.gate_proj.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 254/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.18.mlp.gate_proj.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 254/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.18.mlp.up_proj.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 254/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.18.mlp.up_proj.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 255/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.18.post_attention_layernorm.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 255/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.18.post_attention_layernorm.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 256/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.input_layernorm.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 256/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.input_layernorm.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 257/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.mlp.down_proj.weight] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 257/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.mlp.down_proj.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 258/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.mlp.gate_proj.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 258/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.mlp.gate_proj.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 259/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.mlp.up_proj.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 259/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.mlp.up_proj.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 260/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.post_attention_layernorm.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 260/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.post_attention_layernorm.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 261/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.k_norm.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 261/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.k_norm.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 262/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.k_proj.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 262/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.k_proj.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 263/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.o_proj.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 263/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.o_proj.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 264/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.q_norm.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 264/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.q_norm.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 265/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.q_proj.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 265/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.q_proj.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 266/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.v_proj.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 266/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.19.self_attn.v_proj.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 267/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.input_layernorm.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 267/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.input_layernorm.weight] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 268/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.A_log] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 268/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.A_log] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 269/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.conv1d.weight] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 269/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.conv1d.weight] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 270/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.dt_bias] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 270/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.dt_bias] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 271/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.in_proj_a.weight] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 271/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.in_proj_a.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 272/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.in_proj_b.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 272/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.in_proj_b.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 273/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.in_proj_qkv.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 273/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.in_proj_qkv.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 274/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.in_proj_z.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 274/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.in_proj_z.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 275/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.norm.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 275/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.norm.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 276/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.out_proj.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 276/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.linear_attn.out_proj.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 277/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.mlp.down_proj.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 277/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.mlp.down_proj.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 278/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.mlp.gate_proj.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 278/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.mlp.gate_proj.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 279/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.mlp.up_proj.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 279/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.mlp.up_proj.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 280/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.post_attention_layernorm.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 280/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.20.post_attention_layernorm.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 281/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.input_layernorm.weight] Loading weights: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 281/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.input_layernorm.weight] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 282/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.A_log] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 282/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.A_log] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 283/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.conv1d.weight] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 283/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.conv1d.weight] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 284/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.dt_bias] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 284/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.dt_bias] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 285/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.in_proj_a.weight] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 285/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.in_proj_a.weight] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 286/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.in_proj_b.weight] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 286/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.in_proj_b.weight] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 287/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.in_proj_qkv.weight] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 287/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.in_proj_qkv.weight] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 288/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.in_proj_z.weight] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 288/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.in_proj_z.weight] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 289/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.norm.weight] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 289/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.norm.weight] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 290/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.out_proj.weight] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 290/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.linear_attn.out_proj.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 291/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.mlp.down_proj.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 291/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.mlp.down_proj.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 292/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.mlp.gate_proj.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 292/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.mlp.gate_proj.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 293/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.mlp.up_proj.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 293/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.mlp.up_proj.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 294/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.post_attention_layernorm.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 294/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.21.post_attention_layernorm.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 295/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.22.input_layernorm.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 295/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.22.input_layernorm.weight] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 296/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.22.linear_attn.A_log] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 296/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.22.linear_attn.A_log] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 297/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.22.linear_attn.conv1d.weight] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 297/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.22.linear_attn.conv1d.weight] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 298/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.22.linear_attn.dt_bias] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 298/473 [00:00<00:00, 408.43it/s, Materializing param=language_model.layers.22.linear_attn.dt_bias] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 299/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.dt_bias] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 299/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.in_proj_a.weight] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 299/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.in_proj_a.weight] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 300/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.in_proj_b.weight] Loading weights: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 300/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.in_proj_b.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 301/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.in_proj_qkv.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 301/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.in_proj_qkv.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 302/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.in_proj_z.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 302/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.in_proj_z.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 303/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.norm.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 303/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.norm.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 304/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.out_proj.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 304/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.linear_attn.out_proj.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 305/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.mlp.down_proj.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 305/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.mlp.down_proj.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 306/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.mlp.gate_proj.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 306/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.mlp.gate_proj.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 307/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.mlp.up_proj.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 307/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.mlp.up_proj.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 308/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.post_attention_layernorm.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 308/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.22.post_attention_layernorm.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 309/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.input_layernorm.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 309/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.input_layernorm.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 310/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.mlp.down_proj.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 310/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.mlp.down_proj.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 311/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.mlp.gate_proj.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 311/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.mlp.gate_proj.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 312/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.mlp.up_proj.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 312/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.mlp.up_proj.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 313/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.post_attention_layernorm.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 313/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.post_attention_layernorm.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 314/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.k_norm.weight] Loading weights: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 314/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.k_norm.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 315/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.k_proj.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 315/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.k_proj.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 316/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.o_proj.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 316/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.o_proj.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 317/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.q_norm.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 317/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.q_norm.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 318/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.q_proj.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 318/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.q_proj.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 319/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.v_proj.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 319/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.layers.23.self_attn.v_proj.weight] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 320/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.norm.weight] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 320/473 [00:00<00:00, 366.23it/s, Materializing param=language_model.norm.weight] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 321/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.attn.proj.bias] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 321/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.attn.proj.bias] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 322/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.attn.proj.weight] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 322/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.attn.proj.weight] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 323/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.attn.qkv.bias] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 323/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.attn.qkv.bias] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 324/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.attn.qkv.weight] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 324/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.attn.qkv.weight] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 325/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.mlp.linear_fc1.bias] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 325/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.mlp.linear_fc1.bias] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 326/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.mlp.linear_fc1.weight] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 326/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.mlp.linear_fc1.weight] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 327/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.mlp.linear_fc2.bias] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 327/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.mlp.linear_fc2.bias] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 328/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.mlp.linear_fc2.weight] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 328/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.mlp.linear_fc2.weight] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 329/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.norm1.bias] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 329/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.norm1.bias] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 330/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.norm1.weight] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 330/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.norm1.weight] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 331/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.norm2.bias] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 331/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.norm2.bias] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 332/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.norm2.weight] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 332/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.0.norm2.weight] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 333/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.1.attn.proj.bias] Loading weights: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 333/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.1.attn.proj.bias] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 334/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.1.attn.proj.weight] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 334/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.1.attn.proj.weight] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 335/473 [00:00<00:00, 366.23it/s, Materializing param=visual.blocks.1.attn.qkv.bias] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 335/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.attn.qkv.bias] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 336/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.attn.qkv.weight] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 336/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.attn.qkv.weight] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 337/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.mlp.linear_fc1.bias] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 337/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.mlp.linear_fc1.bias] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 338/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.mlp.linear_fc1.weight] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 338/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.mlp.linear_fc1.weight] Loading weights: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 339/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.mlp.linear_fc2.bias] Loading weights: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 339/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.mlp.linear_fc2.bias] Loading weights: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 340/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.mlp.linear_fc2.weight] Loading weights: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 340/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.mlp.linear_fc2.weight] Loading weights: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 341/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.norm1.bias] Loading weights: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 341/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.norm1.bias] Loading weights: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 342/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.norm1.weight] Loading weights: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 342/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.norm1.weight] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 343/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.norm2.bias] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 343/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.norm2.bias] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 344/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.norm2.weight] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 344/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.1.norm2.weight] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 345/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.attn.proj.bias] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 345/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.attn.proj.bias] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 346/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.attn.proj.weight] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 346/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.attn.proj.weight] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 347/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.attn.qkv.bias] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 347/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.attn.qkv.bias] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 348/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.attn.qkv.weight] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 348/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.attn.qkv.weight] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 349/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.mlp.linear_fc1.bias] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 349/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.mlp.linear_fc1.bias] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 350/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.mlp.linear_fc1.weight] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 350/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.mlp.linear_fc1.weight] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 351/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.mlp.linear_fc2.bias] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 351/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.mlp.linear_fc2.bias] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 352/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.mlp.linear_fc2.weight] Loading weights: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 352/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.mlp.linear_fc2.weight] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 353/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.norm1.bias] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 353/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.norm1.bias] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 354/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.norm1.weight] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 354/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.norm1.weight] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 355/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.norm2.bias] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 355/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.norm2.bias] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 356/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.norm2.weight] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 356/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.2.norm2.weight] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 357/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.attn.proj.bias] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 357/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.attn.proj.bias] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 358/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.attn.proj.weight] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 358/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.attn.proj.weight] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 359/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.attn.qkv.bias] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 359/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.attn.qkv.bias] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 360/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.attn.qkv.weight] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 360/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.attn.qkv.weight] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 361/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.mlp.linear_fc1.bias] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 361/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.mlp.linear_fc1.bias] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 362/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.mlp.linear_fc1.weight] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 362/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.mlp.linear_fc1.weight] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 363/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.mlp.linear_fc2.bias] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 363/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.mlp.linear_fc2.bias] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 364/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.mlp.linear_fc2.weight] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 364/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.mlp.linear_fc2.weight] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 365/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.norm1.bias] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 365/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.norm1.bias] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 366/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.norm1.weight] Loading weights: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 366/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.norm1.weight] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 367/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.norm2.bias] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 367/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.norm2.bias] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 368/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.norm2.weight] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 368/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.3.norm2.weight] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 369/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.attn.proj.bias] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 369/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.attn.proj.bias] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 370/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.attn.proj.weight] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 370/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.attn.proj.weight] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 371/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.attn.qkv.bias] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 371/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.attn.qkv.bias] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 372/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.attn.qkv.weight] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 372/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.attn.qkv.weight] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 373/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.mlp.linear_fc1.bias] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 373/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.mlp.linear_fc1.bias] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 374/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.mlp.linear_fc1.weight] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 374/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.mlp.linear_fc1.weight] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 375/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.mlp.linear_fc2.bias] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 375/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.mlp.linear_fc2.bias] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 376/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.mlp.linear_fc2.weight] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 376/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.mlp.linear_fc2.weight] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 377/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.norm1.bias] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 377/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.norm1.bias] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 378/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.norm1.weight] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 378/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.norm1.weight] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 379/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.norm2.bias] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 379/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.norm2.bias] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 380/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.norm2.weight] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 380/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.4.norm2.weight] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 381/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.attn.proj.bias] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 381/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.attn.proj.bias] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 382/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.attn.proj.weight] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 382/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.attn.proj.weight] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 383/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.attn.qkv.bias] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 383/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.attn.qkv.bias] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 384/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.attn.qkv.weight] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 384/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.attn.qkv.weight] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 385/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.mlp.linear_fc1.bias] Loading weights: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 385/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.mlp.linear_fc1.bias] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 386/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.mlp.linear_fc1.weight] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 386/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.mlp.linear_fc1.weight] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 387/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.mlp.linear_fc2.bias] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 387/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.mlp.linear_fc2.bias] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 388/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.mlp.linear_fc2.weight] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 388/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.mlp.linear_fc2.weight] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 389/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.norm1.bias] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 389/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.norm1.bias] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 390/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.norm1.weight] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 390/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.norm1.weight] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 391/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.norm2.bias] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 391/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.norm2.bias] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 392/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.norm2.weight] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 392/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.5.norm2.weight] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 393/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.attn.proj.bias] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 393/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.attn.proj.bias] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 394/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.attn.proj.weight] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 394/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.attn.proj.weight] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 395/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.attn.qkv.bias] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 395/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.attn.qkv.bias] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 396/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.attn.qkv.weight] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 396/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.attn.qkv.weight] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 397/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.mlp.linear_fc1.bias] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 397/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.mlp.linear_fc1.bias] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 398/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.mlp.linear_fc1.weight] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 398/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.mlp.linear_fc1.weight] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 399/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.mlp.linear_fc2.bias] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 399/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.mlp.linear_fc2.bias] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 400/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.mlp.linear_fc2.weight] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 400/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.mlp.linear_fc2.weight] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 401/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.norm1.bias] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 401/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.norm1.bias] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 402/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.norm1.weight] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 402/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.norm1.weight] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 403/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.norm2.bias] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 403/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.norm2.bias] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 404/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.norm2.weight] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 404/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.6.norm2.weight] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 405/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.attn.proj.bias] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 405/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.attn.proj.bias] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 406/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.attn.proj.weight] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 406/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.attn.proj.weight] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 407/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.attn.qkv.bias] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 407/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.attn.qkv.bias] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 408/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.attn.qkv.weight] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 408/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.attn.qkv.weight] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 409/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.mlp.linear_fc1.bias] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 409/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.mlp.linear_fc1.bias] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 410/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.mlp.linear_fc1.weight] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 410/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.mlp.linear_fc1.weight] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 411/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.mlp.linear_fc2.bias] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 411/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.mlp.linear_fc2.bias] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 412/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.mlp.linear_fc2.weight] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 412/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.mlp.linear_fc2.weight] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 413/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.norm1.bias] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 413/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.norm1.bias] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 414/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.norm1.weight] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 414/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.norm1.weight] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 415/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.norm2.bias] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 415/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.norm2.bias] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 416/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.norm2.weight] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 416/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.7.norm2.weight] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 417/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.8.attn.proj.bias] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 417/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.8.attn.proj.bias] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 418/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.8.attn.proj.weight] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 418/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.8.attn.proj.weight] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 419/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.8.attn.qkv.bias] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 419/473 [00:01<00:00, 366.23it/s, Materializing param=visual.blocks.8.attn.qkv.bias] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 420/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.attn.qkv.bias] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 420/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.attn.qkv.weight] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 420/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.attn.qkv.weight] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 421/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.mlp.linear_fc1.bias] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 421/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.mlp.linear_fc1.bias] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 422/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.mlp.linear_fc1.weight] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 422/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.mlp.linear_fc1.weight] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 423/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.mlp.linear_fc2.bias] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 423/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.mlp.linear_fc2.bias] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 424/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.mlp.linear_fc2.weight] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 424/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.mlp.linear_fc2.weight] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 425/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.norm1.bias] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 425/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.norm1.bias] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 426/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.norm1.weight] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 426/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.norm1.weight] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 427/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.norm2.bias] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 427/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.norm2.bias] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 428/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.norm2.weight] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 428/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.8.norm2.weight] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 429/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.attn.proj.bias] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 429/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.attn.proj.bias] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 430/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.attn.proj.weight] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 430/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.attn.proj.weight] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 431/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.attn.qkv.bias] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 431/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.attn.qkv.bias] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 432/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.attn.qkv.weight] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 432/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.attn.qkv.weight] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 433/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.mlp.linear_fc1.bias] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 433/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.mlp.linear_fc1.bias] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 434/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.mlp.linear_fc1.weight] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 434/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.mlp.linear_fc1.weight] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 435/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.mlp.linear_fc2.bias] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 435/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.mlp.linear_fc2.bias] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 436/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.mlp.linear_fc2.weight] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 436/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.mlp.linear_fc2.weight] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 437/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.norm1.bias] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 437/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.norm1.bias] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 438/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.norm1.weight] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 438/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.norm1.weight] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 439/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.norm2.bias] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 439/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.norm2.bias] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 440/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.norm2.weight] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 440/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.9.norm2.weight] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 441/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.attn.proj.bias] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 441/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.attn.proj.bias] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 442/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.attn.proj.weight] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 442/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.attn.proj.weight] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 443/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.attn.qkv.bias] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 443/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.attn.qkv.bias] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 444/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.attn.qkv.weight] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 444/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.attn.qkv.weight] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 445/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.mlp.linear_fc1.bias] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 445/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.mlp.linear_fc1.bias] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 446/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.mlp.linear_fc1.weight] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 446/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.mlp.linear_fc1.weight] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 447/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.mlp.linear_fc2.bias] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 447/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.mlp.linear_fc2.bias] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 448/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.mlp.linear_fc2.weight] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 448/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.mlp.linear_fc2.weight] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 449/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.norm1.bias] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 449/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.norm1.bias] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 450/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.norm1.weight] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 450/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.norm1.weight] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 451/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.norm2.bias] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 451/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.norm2.bias] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 452/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.norm2.weight] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 452/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.10.norm2.weight] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 453/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.attn.proj.bias] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 453/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.attn.proj.bias] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 454/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.attn.proj.weight] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 454/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.attn.proj.weight] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 455/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.attn.qkv.bias] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 455/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.attn.qkv.bias] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 456/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.attn.qkv.weight] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 456/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.attn.qkv.weight] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 457/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.mlp.linear_fc1.bias] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 457/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.mlp.linear_fc1.bias] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 458/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.mlp.linear_fc1.weight] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 458/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.mlp.linear_fc1.weight] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 459/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.mlp.linear_fc2.bias] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 459/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.mlp.linear_fc2.bias] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 460/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.mlp.linear_fc2.weight] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 460/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.mlp.linear_fc2.weight] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 461/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.norm1.bias] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 461/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.norm1.bias] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 462/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.norm1.weight] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 462/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.norm1.weight] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 463/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.norm2.bias] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 463/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.norm2.bias] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 464/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.norm2.weight] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 464/473 [00:01<00:00, 594.62it/s, Materializing param=visual.blocks.11.norm2.weight] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 465/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.linear_fc1.bias] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 465/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.linear_fc1.bias] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 466/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.linear_fc1.weight] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 466/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.linear_fc1.weight] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 467/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.linear_fc2.bias] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 467/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.linear_fc2.bias] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 468/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.linear_fc2.weight] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 468/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.linear_fc2.weight] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 469/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.norm.bias] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 469/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.norm.bias] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 470/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.norm.weight] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 470/473 [00:01<00:00, 594.62it/s, Materializing param=visual.merger.norm.weight] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 471/473 [00:01<00:00, 594.62it/s, Materializing param=visual.patch_embed.proj.bias] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 471/473 [00:01<00:00, 594.62it/s, Materializing param=visual.patch_embed.proj.bias] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 472/473 [00:01<00:00, 594.62it/s, Materializing param=visual.patch_embed.proj.weight] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 472/473 [00:01<00:00, 594.62it/s, Materializing param=visual.patch_embed.proj.weight] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 473/473 [00:01<00:00, 594.62it/s, Materializing param=visual.pos_embed.weight] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 473/473 [00:01<00:00, 594.62it/s, Materializing param=visual.pos_embed.weight] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 473/473 [00:01<00:00, 449.91it/s, Materializing param=visual.pos_embed.weight]
Teacher loaded: 853.0M params
Loading student from MVP/artifacts/qwen35-0.8b-text-n4...
Loading weights: 0%| | 0/55 [00:00<?, ?it/s] Loading weights: 2%|▏ | 1/55 [00:00<00:00, 30393.51it/s, Materializing param=model.embed_tokens.weight] Loading weights: 2%|▏ | 1/55 [00:00<00:00, 6563.86it/s, Materializing param=model.embed_tokens.weight] Loading weights: 4%|β–Ž | 2/55 [00:00<00:09, 5.60it/s, Materializing param=model.embed_tokens.weight] Loading weights: 4%|β–Ž | 2/55 [00:00<00:09, 5.60it/s, Materializing param=model.layers.0.input_layernorm.weight] Loading weights: 4%|β–Ž | 2/55 [00:00<00:09, 5.60it/s, Materializing param=model.layers.0.input_layernorm.weight] Loading weights: 5%|β–Œ | 3/55 [00:00<00:09, 5.60it/s, Materializing param=model.layers.0.linear_attn.A_log] Loading weights: 5%|β–Œ | 3/55 [00:00<00:09, 5.60it/s, Materializing param=model.layers.0.linear_attn.A_log] Loading weights: 7%|β–‹ | 4/55 [00:00<00:09, 5.60it/s, Materializing param=model.layers.0.linear_attn.conv1d.weight] Loading weights: 7%|β–‹ | 4/55 [00:00<00:09, 5.60it/s, Materializing param=model.layers.0.linear_attn.conv1d.weight] Loading weights: 9%|β–‰ | 5/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.dt_bias] Loading weights: 9%|β–‰ | 5/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.dt_bias] Loading weights: 11%|β–ˆ | 6/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.in_proj_a.weight] Loading weights: 11%|β–ˆ | 6/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.in_proj_a.weight] Loading weights: 13%|β–ˆβ–Ž | 7/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.in_proj_b.weight] Loading weights: 13%|β–ˆβ–Ž | 7/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.in_proj_b.weight] Loading weights: 15%|β–ˆβ– | 8/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.in_proj_qkv.weight] Loading weights: 15%|β–ˆβ– | 8/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.in_proj_qkv.weight] Loading weights: 16%|β–ˆβ–‹ | 9/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.in_proj_z.weight] Loading weights: 16%|β–ˆβ–‹ | 9/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.in_proj_z.weight] Loading weights: 18%|β–ˆβ–Š | 10/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.norm.weight] Loading weights: 18%|β–ˆβ–Š | 10/55 [00:00<00:08, 5.60it/s, Materializing param=model.layers.0.linear_attn.norm.weight] Loading weights: 20%|β–ˆβ–ˆ | 11/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.linear_attn.out_proj.weight] Loading weights: 20%|β–ˆβ–ˆ | 11/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.linear_attn.out_proj.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 12/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.mlp.down_proj.weight] Loading weights: 22%|β–ˆβ–ˆβ– | 12/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.mlp.down_proj.weight] Loading weights: 24%|β–ˆβ–ˆβ–Ž | 13/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.mlp.gate_proj.weight] Loading weights: 24%|β–ˆβ–ˆβ–Ž | 13/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.mlp.gate_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ–Œ | 14/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.mlp.up_proj.weight] Loading weights: 25%|β–ˆβ–ˆβ–Œ | 14/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.mlp.up_proj.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 15/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.post_attention_layernorm.weight] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 15/55 [00:00<00:07, 5.60it/s, Materializing param=model.layers.0.post_attention_layernorm.weight] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 16/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.input_layernorm.weight] Loading weights: 29%|β–ˆβ–ˆβ–‰ | 16/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.input_layernorm.weight] Loading weights: 31%|β–ˆβ–ˆβ–ˆ | 17/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.A_log] Loading weights: 31%|β–ˆβ–ˆβ–ˆ | 17/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.A_log] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 18/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.conv1d.weight] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 18/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.conv1d.weight] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ– | 19/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.dt_bias] Loading weights: 35%|β–ˆβ–ˆβ–ˆβ– | 19/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.dt_bias] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–‹ | 20/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.in_proj_a.weight] Loading weights: 36%|β–ˆβ–ˆβ–ˆβ–‹ | 20/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.in_proj_a.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 21/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.in_proj_b.weight] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 21/55 [00:00<00:06, 5.60it/s, Materializing param=model.layers.1.linear_attn.in_proj_b.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 22/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.linear_attn.in_proj_qkv.weight] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 22/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.linear_attn.in_proj_qkv.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 23/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.linear_attn.in_proj_z.weight] Loading weights: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 23/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.linear_attn.in_proj_z.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 24/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.linear_attn.norm.weight] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 24/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.linear_attn.norm.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 25/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.linear_attn.out_proj.weight] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 25/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.linear_attn.out_proj.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 26/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.mlp.down_proj.weight] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 26/55 [00:00<00:05, 5.60it/s, Materializing param=model.layers.1.mlp.down_proj.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 27/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.1.mlp.gate_proj.weight] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 27/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.1.mlp.gate_proj.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 28/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.1.mlp.up_proj.weight] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 28/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.1.mlp.up_proj.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 29/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.1.post_attention_layernorm.weight] Loading weights: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 29/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.1.post_attention_layernorm.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 30/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.2.input_layernorm.weight] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 30/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.2.input_layernorm.weight] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 31/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.2.linear_attn.A_log] Loading weights: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 31/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.2.linear_attn.A_log] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 32/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.2.linear_attn.conv1d.weight] Loading weights: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 32/55 [00:00<00:04, 5.60it/s, Materializing param=model.layers.2.linear_attn.conv1d.weight] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 33/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.dt_bias] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 33/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.dt_bias] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 34/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.in_proj_a.weight] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 34/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.in_proj_a.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 35/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.in_proj_b.weight] Loading weights: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 35/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.in_proj_b.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 36/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.in_proj_qkv.weight] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 36/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.in_proj_qkv.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 37/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.in_proj_z.weight] Loading weights: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 37/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.in_proj_z.weight] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 38/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.norm.weight] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 38/55 [00:00<00:03, 5.60it/s, Materializing param=model.layers.2.linear_attn.norm.weight] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 39/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.linear_attn.out_proj.weight] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 39/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.linear_attn.out_proj.weight] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 40/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.mlp.down_proj.weight] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 40/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.mlp.down_proj.weight] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 41/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.mlp.gate_proj.weight] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 41/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.mlp.gate_proj.weight] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 42/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.mlp.up_proj.weight] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 42/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.mlp.up_proj.weight] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 43/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.post_attention_layernorm.weight] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 43/55 [00:00<00:02, 5.60it/s, Materializing param=model.layers.2.post_attention_layernorm.weight] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 44/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.input_layernorm.weight] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 44/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.input_layernorm.weight] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 45/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.mlp.down_proj.weight] Loading weights: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 45/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.mlp.down_proj.weight] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 46/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.mlp.gate_proj.weight] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 46/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.mlp.gate_proj.weight] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 47/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.mlp.up_proj.weight] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 47/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.mlp.up_proj.weight] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 48/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.post_attention_layernorm.weight] Loading weights: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 48/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.post_attention_layernorm.weight] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 49/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.self_attn.k_norm.weight] Loading weights: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 49/55 [00:00<00:01, 5.60it/s, Materializing param=model.layers.3.self_attn.k_norm.weight] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 50/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.k_proj.weight] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 50/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.k_proj.weight] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 51/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.o_proj.weight] Loading weights: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 51/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.o_proj.weight] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 52/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.q_norm.weight] Loading weights: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 52/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.q_norm.weight] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 53/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.q_proj.weight] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 53/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.q_proj.weight] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 54/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.v_proj.weight] Loading weights: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 54/55 [00:00<00:00, 5.60it/s, Materializing param=model.layers.3.self_attn.v_proj.weight] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 55/55 [00:00<00:00, 5.60it/s, Materializing param=model.norm.weight] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 55/55 [00:00<00:00, 5.60it/s, Materializing param=model.norm.weight] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 55/55 [00:00<00:00, 149.22it/s, Materializing param=model.norm.weight]
The tokenizer you are loading from 'MVP/artifacts/qwen35-0.8b-text-n4' with an incorrect regex pattern: https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Instruct-2503/discussions/84#69121093e8b480e709447d5e. This will lead to incorrect tokenization. You should set the `fix_mistral_regex=True` flag when loading this tokenizer to fix this issue.
Token indices sequence length is longer than the specified maximum sequence length for this model (2505798 > Student loaded: 337.3M params
Loading tokenizer...
Tokenizing dataset (/tmp/wik/usr/lib/python3.14/multiprocessing/resource_tracker.py:475: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown: {'/mp-z_hczt5i'}
warnings.warn(
en=128
Step 25/2000 | total=288.4217 ce=13.6517 mse=27.477000 | lr=1.50e-05 | 107s elapsed, ~8492s remaining
Step 50/2000 | total=249.2028 ce=12.1995 mse=23.700326 | lr=3.00e-05 | 214s elapsed, ~8336s remaining
Step 75/2000 | total=182.6658 ce=9.3742 mse=17.329160 | lr=4.50e-05 | 320s elapsed, ~8217s remaining
Step 100/2000 | total=138.5065 ce=7.9229 mse=13.058353 | lr=6.00e-05 | 427s elapsed, ~8108s remaining
Step 125/2000 | total=115.7030 ce=7.5666 mse=10.813636 | lr=7.50e-05 | 532s elapsed, ~7977s remaining
Step 150/2000 | total=106.5914 ce=7.3182 mse=9.927319 | lr=9.00e-05 | 638s elapsed, ~7865s remaining
Step 175/2000 | total=100.8314 ce=6.9378 mse=9.389356 | lr=1.05e-04 | 743s elapsed, ~7747s remaining
Step 200/2000 | total=98.4073 ce=6.5909 mse=9.181641 | lr=1.25e-04 | 850s elapsed, ~7648s remaining
Step 225/2000 | total=93.9046 ce=6.3370 mse=8.756758 | lr=1.40e-04 | 955s elapsed, ~7536s remaining
Step 250/2000 | total=88.4071 ce=6.0484 mse=8.235866 | lr=1.55e-04 | 1539s elapsed, ~10774s remaining
Traceback (most recent call last):
File "/home/void0x14/Documents/echo/MVP/distill.py", line 289, in <module>
_main()
~~~~~^^
File "/home/void0x14/Documents/echo/MVP/distill.py", line 285, in _main
run_distill(DistillConfig(**{k.replace("-", "_"): v for k, v in vars(args).items()}))
~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/void0x14/Documents/echo/MVP/distill.py", line 256, in run_distill
save_checkpoint(student, student_config, args.output_dir, step + 1)
~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/void0x14/Documents/echo/MVP/distill.py", line 160, in save_checkpoint
config_path.write_text(json.dumps(student_config.to_dict(), indent=2, sort_keys=True) + "\n", encoding="utf-8")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.14/json/__init__.py", line 242, in dumps
**kw).encode(obj)
~~~~~~^^^^^
File "/usr/lib/python3.14/json/encoder.py", line 202, in encode
chunks = self.iterencode(o, _one_shot=True)
File "/usr/lib/python3.14/json/encoder.py", line 263, in iterencode
return _iterencode(o, 0)
File "/usr/lib/python3.14/json/encoder.py", line 182, in default
raise TypeError(f'Object of type {o.__class__.__name__} '
f'is not JSON serializable')
TypeError: Object of type Qwen3_5TextConfig is not JSON serializable