Issues
- 0
训练数据中指令与模型行为不匹配
#58 opened by haichao592 - 0
本地模型
#57 opened by lz2021211161 - 1
请问哪里可以找到工作里对于数据库方面的训练数据
#56 opened by Mucalinda2436 - 1
weight decay确定是0.1吗?
#54 opened by Fu-Dayuan - 0
魔塔上的 AgentInstruct 数据集的 conversation 都是空值
#55 opened by XianglongTan - 3
基于fastchat部署,推理异常
#46 opened by ruifengma - 1
貌似hotpotqa测试脚本跑不起来?
#53 opened by Fu-Dayuan - 3
训练数据是如何采样的?
#52 opened by Fu-Dayuan - 7
- 1
- 1
if it is possible to conduct RLHF from env
#51 opened by SHITIANYU-hue - 1
- 1
可以给个简单点的工具调用示例吗
#48 opened by qq594495953 - 1
期待用 Qwen72B 训练的模型。
#47 opened by milomoon - 6
除了用docker运行,还有其他方式可以运行AgentLM吗?
#42 opened by caizhuoyue77 - 3
关于TRAJECTORY FILTERING问题
#44 opened by QingChengLineOne - 13
- 5
请问下agentlm-7b最少需要多少显存可以推理
#45 opened by nicolasNi - 7
论文中关于损失函数E的问题
#10 opened by Dhaizei - 0
Adding Contributors Section in readme.md file.
#27 opened by mohitd404 - 1
Finetuning with Mistral or Yi?
#43 opened by jFkd1 - 0
关于数据集
#37 opened by DryPilgrim - 1
Number of training steps
#36 opened by Mayer123 - 5
Dataset details 中找不到reward的计算方式
#40 opened by DryPilgrim - 3
关于dataset statics 和 download
#38 opened by DryPilgrim - 2
关于reward
#32 opened by DryPilgrim - 1
agent tuning和toolbench的区别
#34 opened by Connor-Shen - 1
微调显存
#35 opened by Reason-Wang - 1
Start TGI worker
#33 opened by mayilin0714 - 1
请教reward分数的各种情况
#30 opened by DryPilgrim - 1
requests.exceptions.MissingSchema: Invalid URL '127.0.0.123332/generate': No scheme supplied. Perhaps you meant https://127.0.0.123332/generate?
#31 opened by mayilin0714 - 1
Inference with `vllm`
#29 opened by yc1999 - 2
论文中Table 2中的数字的含义和计算方式
#17 opened by DryPilgrim - 0
Add license
#24 opened by dalvishruti14 - 1
Grammer mistake in readme
#19 opened by shraddha761 - 1
Auto comment
#21 opened by shraddha761 - 1
什么时候上魔塔社区
#18 opened by QingChengLineOne - 1
论文中的问题
#16 opened by QingChengLineOne - 2
Model Output Length
#11 opened by HeimingX - 2
是否可以不在docker里运行
#5 opened by QingChengLineOne - 1
底座模型基于llama2,是否支持中文呢
#4 opened by cnsky2016 - 1
fine-tune code
#12 opened by faaany - 2
- 0
- 1
交互轨迹的Reward如何得到
#8 opened by jiezhangGt - 1
AgentLM能支持openai.api类的接口本地部署吗?
#3 opened by HuntZhaozq - 3
train code
#1 opened by wangyanli3630