Comments (4)
zhaochenyang20
commented on May 30, 2024
14
历史级的笑话 🤣
有啥好笑的，这是base模型，又不是chat模型，这种输出很正常
从一个预训练的角度来看，这个模型至少存在两个问题。首先是 model collapse，也即还在循环输出看上去连贯的内容。其次，更本质和严重的是，这个模型已经完全过拟合了。为了拟合某些 benchmark（实际上是 C-Eval），这个模型已经疯狂到了题目都没看到，就开始做题的地步了。我想了想用什么来回复，又或者是说如何回复现在国内的模型厂家。我的回复大概率是....
from yi.
zhaochenyang20
commented on May 30, 2024
1
历史级的笑话 🤣
from yi.
hanyi-zou
commented on May 30, 2024
历史级的笑话 🤣
有啥好笑的，这是base模型，又不是chat模型，这种输出很正常
from yi.
zheng5yu9
commented on May 30, 2024
单一个例不能说明什么，要有系统的使用评测，能够放出来说明有点信心。还有好多不开源的呢
from yi.
Related Issues (20)
- 就不能在官网挂一个体验的链接吗 HOT 1
- 6b运行正常，34b-int4运行失败 HOT 3
- 用示例代码，返回值不符合预期。 HOT 1
- 200K 上下文的模型什么时候能放出 chat 模型呢？ HOT 4
- How to inference in a multi-batch way? HOT 1
- 运行Yi-34B-Chat-4bits内核报错 HOT 1
- 用 VLLM 加载 Yi-34B-Chat-4bits-GPTQ, 模型有时不停地输出空字串而不停止 HOT 7
- 请问yi-34B-chat的模型支持更长的上下文吗？ HOT 1
- RuntimeError: probability tensor contains either `inf`, `nan` or element < 0 HOT 1
- yi-34b预训练模型支持多GPU分布参数进行量化和微调吗？ HOT 1
- The train/eval data format is different between the finetune guide and the chat model's chat template HOT 2
- sft报错：ValueError: YiForCausalLM does not support Flash Attention 2.0 yet. HOT 5
- 请教微调Yi-34B-Chat模型时context长度问题 HOT 1
- 微调 YI-6B 一直出现 loss scale overflow 然后 reduce 到 min_loss_scale 报错, YI-6B-Chat 则没问题，chat 模型训练参数设置有什么不同吗 HOT 4
- Function Calling 功能支持计划
- 参考finetune的quickstart，但模型文件中没有modeling_yi.py啊 HOT 3
- Yi
- Yi 支持LongLoRA吗？ HOT 1
- RuntimeError: You can't move a model that has some modules offloaded to cpu or disk. HOT 3
- Yi-34B-Chat-4bit 无法配合langchain作为agent使用 HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from yi.