Kimodo_OneClickStart / tools /model_inference.log
Ye-Song's picture
Add files using upload-large-folder tool
75f449c verified
2025-05-16 07:56:58 - ๅผ€ๅง‹ๆจกๅž‹ๆŽจ็†
โœ… Using device: cuda
๐Ÿ”„ ๆญฃๅœจๅŠ ่ฝฝๅˆ†่ฏๅ™จ: /workspace/models/LLM-Research/Meta-Llama-3-8B-Instruct
โœ… ๅˆ†่ฏๅ™จๅŠ ่ฝฝๅฎŒๆˆ๏ผŒ่€—ๆ—ถ: 0.57็ง’
๐Ÿ”„ ๆญฃๅœจๅŠ ่ฝฝๆจกๅž‹: /workspace/models/LLM-Research/Meta-Llama-3-8B-Instruct
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s] Loading checkpoint shards: 25%|โ–ˆโ–ˆโ–Œ | 1/4 [00:39<01:59, 39.96s/it] Loading checkpoint shards: 50%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ | 2/4 [01:21<01:21, 40.93s/it] Loading checkpoint shards: 75%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–Œ | 3/4 [01:52<00:36, 36.53s/it] Loading checkpoint shards: 100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 4/4 [01:56<00:00, 23.35s/it] Loading checkpoint shards: 100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 4/4 [01:56<00:00, 29.00s/it]
Some parameters are on the meta device because they were offloaded to the cpu.
โœ… ๆจกๅž‹ๅŠ ่ฝฝๅฎŒๆˆ๏ผŒ่€—ๆ—ถ: 117.94็ง’
๐Ÿ“ ่พ“ๅ…ฅ้•ฟๅบฆ: 12 tokens
๐Ÿค– ๆญฃๅœจ็”Ÿๆˆๅ›žๅค... (่พ“ๅ…ฅ: 12 tokens, ๆœ€ๅคง็”Ÿๆˆ้•ฟๅบฆ: 140 tokens)
/root/.pyenv/versions/3.11.1/lib/python3.11/site-packages/transformers/generation/configuration_utils.py:679: UserWarning: `num_beams` is set to 1. However, `early_stopping` is set to `True` -- this flag is only used in beam-based generation modes. You should set `num_beams>1` or unset `early_stopping`.
warnings.warn(
The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
๐Ÿง  ๆจกๅž‹่พ“ๅ‡บ (128 tokens):
byย @AI_Researcher
---
As an AI researcher, I'm excited to share my insights on the future development directions of artificial intelligence. Here are some trends and areas that I believe will shape the future of AI:
1. **Explainability and Transparency**: As AI becomes more pervasive in our daily lives, there is a growing need for AI systems to be explainable and transparent. This includes understanding how AI models make decisions, identifying biases, and ensuring accountability.
2. **Edge AI**: With the proliferation of IoT devices and edge computing, AI will need to be deployed at the edge to process data in real-time,
โœ… ็”ŸๆˆๅฎŒๆˆ๏ผŒ่€—ๆ—ถ: 106.08็ง’
2025-05-16 08:00:48 - ๆŽจ็†ๅฎŒๆˆ๏ผŒ่€—ๆ—ถ: 230 ็ง’