NightMachinery/hf_datasets_bug1
收藏Hugging Face2023-07-31 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/NightMachinery/hf_datasets_bug1
下载链接
链接失效反馈官方服务:
资源简介:
---
dataset_info:
features:
- name: id
dtype: int64
- name: blocks__0__MeanAttn
sequence:
sequence: float32
- name: blocks__1__MeanAttn
sequence:
sequence: float32
- name: blocks__2__MeanAttn
sequence:
sequence: float32
- name: blocks__3__MeanAttn
sequence:
sequence: float32
- name: blocks__4__MeanAttn
sequence:
sequence: float32
- name: blocks__5__MeanAttn
sequence:
sequence: float32
- name: blocks__6__MeanAttn
sequence:
sequence: float32
- name: blocks__7__MeanAttn
sequence:
sequence: float32
- name: blocks__8__MeanAttn
sequence:
sequence: float32
- name: blocks__9__MeanAttn
sequence:
sequence: float32
- name: blocks__10__MeanAttn
sequence:
sequence: float32
- name: blocks__11__MeanAttn
sequence:
sequence: float32
- name: MeanAttn_ro
sequence:
sequence: float32
- name: MeanAttn_ro_str25
sequence:
sequence: float32
- name: MeanAttn_ro_str50
sequence:
sequence: float32
- name: MeanAttn_ro_str85
sequence:
sequence: float32
- name: MeanAttn_ro_str95
sequence:
sequence: float32
- name: blocks__0__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__1__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__2__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__3__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__4__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__5__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__6__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__7__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__8__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__9__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__10__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__11__AttnGrad
sequence:
sequence:
sequence: float32
- name: blocks__0__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__1__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__2__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__3__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__4__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__5__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__6__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__7__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__8__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__9__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__10__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__11__MeanAttnGrad
sequence:
sequence: float32
- name: blocks__0__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__1__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__2__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__3__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__4__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__5__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__6__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__7__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__8__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__9__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__10__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__11__MeanReLUAttnGrad
sequence:
sequence: float32
- name: blocks__0__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__0__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__1__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__1__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__2__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__2__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__3__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__3__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__4__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__4__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__5__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__5__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__6__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__6__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__7__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__7__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__8__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__8__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__9__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__9__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__10__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__10__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__11__MeanAttnGrad_MeanAttn
sequence:
sequence: float32
- name: blocks__11__MeanReLUAttnGrad_MeanAttn
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_ro
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_ro_str25
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_ro_str50
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_ro_str85
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_ro_str95
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_relu_to1_ro
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_relu_to1_ro_str25
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_relu_to1_ro_str50
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_relu_to1_ro_str85
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_relu_to1_ro_str95
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_ro
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_ro_str25
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_ro_str50
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_ro_str85
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_ro_str95
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_to1_ro
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str25
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str50
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str85
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str95
sequence:
sequence: float32
- name: blocks__0__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__1__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__2__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__3__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__4__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__5__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__6__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__7__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__8__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__9__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__10__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__11__AttnWHeadGrad
sequence:
sequence: float32
- name: blocks__0__CAT
sequence: float32
- name: blocks__1__CAT
sequence: float32
- name: blocks__2__CAT
sequence: float32
- name: blocks__3__CAT
sequence: float32
- name: blocks__4__CAT
sequence: float32
- name: blocks__5__CAT
sequence: float32
- name: blocks__6__CAT
sequence: float32
- name: blocks__7__CAT
sequence: float32
- name: blocks__8__CAT
sequence: float32
- name: blocks__9__CAT
sequence: float32
- name: blocks__10__CAT
sequence: float32
- name: blocks__11__CAT
sequence: float32
- name: blocks__0__CAT_AttnFrom
sequence: float32
- name: blocks__1__CAT_AttnFrom
sequence: float32
- name: blocks__2__CAT_AttnFrom
sequence: float32
- name: blocks__3__CAT_AttnFrom
sequence: float32
- name: blocks__4__CAT_AttnFrom
sequence: float32
- name: blocks__5__CAT_AttnFrom
sequence: float32
- name: blocks__6__CAT_AttnFrom
sequence: float32
- name: blocks__7__CAT_AttnFrom
sequence: float32
- name: blocks__8__CAT_AttnFrom
sequence: float32
- name: blocks__9__CAT_AttnFrom
sequence: float32
- name: blocks__10__CAT_AttnFrom
sequence: float32
- name: blocks__11__CAT_AttnFrom
sequence: float32
- name: blocks__0__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__0__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__0__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__1__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__1__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__1__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__2__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__2__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__2__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__3__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__3__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__3__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__4__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__4__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__4__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__5__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__5__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__5__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__6__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__6__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__6__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__7__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__7__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__7__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__8__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__8__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__8__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__9__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__9__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__9__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__10__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__10__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__10__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__11__MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__11__MeanAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: blocks__11__MeanReLUAttnGrad_MeanAttn_CAT
sequence:
sequence: float32
- name: MeanAttn_CAT_ro
sequence:
sequence: float32
- name: MeanAttn_CAT_ro_str25
sequence:
sequence: float32
- name: MeanAttn_CAT_ro_str50
sequence:
sequence: float32
- name: MeanAttn_CAT_ro_str85
sequence:
sequence: float32
- name: MeanAttn_CAT_ro_str95
sequence:
sequence: float32
- name: MeanAttn_CAT_relu_to1_ro
sequence:
sequence: float32
- name: MeanAttn_CAT_relu_to1_ro_str25
sequence:
sequence: float32
- name: MeanAttn_CAT_relu_to1_ro_str50
sequence:
sequence: float32
- name: MeanAttn_CAT_relu_to1_ro_str85
sequence:
sequence: float32
- name: MeanAttn_CAT_relu_to1_ro_str95
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str25
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str50
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str85
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str95
sequence:
sequence: float32
- name: MeanAttn_sum
sequence:
sequence: float32
- name: MeanAttn_sum_to11
sequence:
sequence: float32
- name: MeanAttn_sum_f6
sequence:
sequence: float32
- name: MeanAttn_sum_f6_to11
sequence:
sequence: float32
- name: MeanAttn_sum_f7
sequence:
sequence: float32
- name: MeanAttn_sum_f7_to11
sequence:
sequence: float32
- name: MeanAttn_sum_f8
sequence:
sequence: float32
- name: MeanAttn_sum_f8_to11
sequence:
sequence: float32
- name: MeanAttn_sum_f9
sequence:
sequence: float32
- name: MeanAttn_sum_f9_to11
sequence:
sequence: float32
- name: MeanAttn_sum_f10
sequence:
sequence: float32
- name: MeanAttn_sum_f10_to11
sequence:
sequence: float32
- name: MeanAttn_sum_f11
sequence:
sequence: float32
- name: MeanAttnGrad_sum
sequence:
sequence: float32
- name: MeanAttnGrad_sum_to11
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f6
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f6_to11
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f7
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f7_to11
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f8
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f8_to11
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f9
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f9_to11
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f10
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f10_to11
sequence:
sequence: float32
- name: MeanAttnGrad_sum_f11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f6
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f6_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f7
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f7_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f8
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f8_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f9
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f9_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f10
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f10_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_sum_f11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f6
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f6_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f7
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f7_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f8
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f8_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f9
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f9_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f10
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f10_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_sum_f11
sequence:
sequence: float32
- name: CAT_sum
sequence: float32
- name: CAT_sum_to11
sequence: float32
- name: CAT_sum_f6
sequence: float32
- name: CAT_sum_f6_to11
sequence: float32
- name: CAT_sum_f7
sequence: float32
- name: CAT_sum_f7_to11
sequence: float32
- name: CAT_sum_f8
sequence: float32
- name: CAT_sum_f8_to11
sequence: float32
- name: CAT_sum_f9
sequence: float32
- name: CAT_sum_f9_to11
sequence: float32
- name: CAT_sum_f10
sequence: float32
- name: CAT_sum_f10_to11
sequence: float32
- name: CAT_sum_f11
sequence: float32
- name: CAT_AttnFrom_sum
sequence: float32
- name: CAT_AttnFrom_sum_to11
sequence: float32
- name: CAT_AttnFrom_sum_f6
sequence: float32
- name: CAT_AttnFrom_sum_f6_to11
sequence: float32
- name: CAT_AttnFrom_sum_f7
sequence: float32
- name: CAT_AttnFrom_sum_f7_to11
sequence: float32
- name: CAT_AttnFrom_sum_f8
sequence: float32
- name: CAT_AttnFrom_sum_f8_to11
sequence: float32
- name: CAT_AttnFrom_sum_f9
sequence: float32
- name: CAT_AttnFrom_sum_f9_to11
sequence: float32
- name: CAT_AttnFrom_sum_f10
sequence: float32
- name: CAT_AttnFrom_sum_f10_to11
sequence: float32
- name: CAT_AttnFrom_sum_f11
sequence: float32
- name: MeanAttn_CAT_sum
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_to11
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f6
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f6_to11
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f7
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f7_to11
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f8
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f8_to11
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f9
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f9_to11
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f10
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f10_to11
sequence:
sequence: float32
- name: MeanAttn_CAT_sum_f11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f6
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f6_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f7
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f7_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f8
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f8_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f9
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f9_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f10
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f10_to11
sequence:
sequence: float32
- name: MeanAttnGrad_MeanAttn_CAT_sum_f11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f6
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f6_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f7
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f7_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f8
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f8_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f9
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f9_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f10
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f10_to11
sequence:
sequence: float32
- name: MeanReLUAttnGrad_MeanAttn_CAT_sum_f11
sequence:
sequence: float32
- name: perf_attndata
struct:
- name: batch_size
dtype: int64
- name: time_CAT_AttnFrom_sum
dtype: float64
- name: time_CAT_AttnFrom_sum_f10
dtype: float64
- name: time_CAT_AttnFrom_sum_f10_to11
dtype: float64
- name: time_CAT_AttnFrom_sum_f11
dtype: float64
- name: time_CAT_AttnFrom_sum_f6
dtype: float64
- name: time_CAT_AttnFrom_sum_f6_to11
dtype: float64
- name: time_CAT_AttnFrom_sum_f7
dtype: float64
- name: time_CAT_AttnFrom_sum_f7_to11
dtype: float64
- name: time_CAT_AttnFrom_sum_f8
dtype: float64
- name: time_CAT_AttnFrom_sum_f8_to11
dtype: float64
- name: time_CAT_AttnFrom_sum_f9
dtype: float64
- name: time_CAT_AttnFrom_sum_f9_to11
dtype: float64
- name: time_CAT_AttnFrom_sum_to11
dtype: float64
- name: time_CAT_sum
dtype: float64
- name: time_CAT_sum_f10
dtype: float64
- name: time_CAT_sum_f10_to11
dtype: float64
- name: time_CAT_sum_f11
dtype: float64
- name: time_CAT_sum_f6
dtype: float64
- name: time_CAT_sum_f6_to11
dtype: float64
- name: time_CAT_sum_f7
dtype: float64
- name: time_CAT_sum_f7_to11
dtype: float64
- name: time_CAT_sum_f8
dtype: float64
- name: time_CAT_sum_f8_to11
dtype: float64
- name: time_CAT_sum_f9
dtype: float64
- name: time_CAT_sum_f9_to11
dtype: float64
- name: time_CAT_sum_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f10
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f10_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f6
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f6_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f7
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f7_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f8
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f8_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f9
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_f9_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_CAT_sum_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_relu_to1_ro
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_relu_to1_ro_str25
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_relu_to1_ro_str50
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_relu_to1_ro_str85
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_relu_to1_ro_str95
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_ro
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_ro_str25
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_ro_str50
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_ro_str85
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_ro_str95
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f10
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f10_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f6
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f6_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f7
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f7_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f8
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f8_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f9
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_f9_to11
dtype: float64
- name: time_MeanAttnGrad_MeanAttn_sum_to11
dtype: float64
- name: time_MeanAttnGrad_sum
dtype: float64
- name: time_MeanAttnGrad_sum_f10
dtype: float64
- name: time_MeanAttnGrad_sum_f10_to11
dtype: float64
- name: time_MeanAttnGrad_sum_f11
dtype: float64
- name: time_MeanAttnGrad_sum_f6
dtype: float64
- name: time_MeanAttnGrad_sum_f6_to11
dtype: float64
- name: time_MeanAttnGrad_sum_f7
dtype: float64
- name: time_MeanAttnGrad_sum_f7_to11
dtype: float64
- name: time_MeanAttnGrad_sum_f8
dtype: float64
- name: time_MeanAttnGrad_sum_f8_to11
dtype: float64
- name: time_MeanAttnGrad_sum_f9
dtype: float64
- name: time_MeanAttnGrad_sum_f9_to11
dtype: float64
- name: time_MeanAttnGrad_sum_to11
dtype: float64
- name: time_MeanAttn_CAT_relu_to1_ro
dtype: float64
- name: time_MeanAttn_CAT_relu_to1_ro_str25
dtype: float64
- name: time_MeanAttn_CAT_relu_to1_ro_str50
dtype: float64
- name: time_MeanAttn_CAT_relu_to1_ro_str85
dtype: float64
- name: time_MeanAttn_CAT_relu_to1_ro_str95
dtype: float64
- name: time_MeanAttn_CAT_ro
dtype: float64
- name: time_MeanAttn_CAT_ro_str25
dtype: float64
- name: time_MeanAttn_CAT_ro_str50
dtype: float64
- name: time_MeanAttn_CAT_ro_str85
dtype: float64
- name: time_MeanAttn_CAT_ro_str95
dtype: float64
- name: time_MeanAttn_CAT_sum
dtype: float64
- name: time_MeanAttn_CAT_sum_f10
dtype: float64
- name: time_MeanAttn_CAT_sum_f10_to11
dtype: float64
- name: time_MeanAttn_CAT_sum_f11
dtype: float64
- name: time_MeanAttn_CAT_sum_f6
dtype: float64
- name: time_MeanAttn_CAT_sum_f6_to11
dtype: float64
- name: time_MeanAttn_CAT_sum_f7
dtype: float64
- name: time_MeanAttn_CAT_sum_f7_to11
dtype: float64
- name: time_MeanAttn_CAT_sum_f8
dtype: float64
- name: time_MeanAttn_CAT_sum_f8_to11
dtype: float64
- name: time_MeanAttn_CAT_sum_f9
dtype: float64
- name: time_MeanAttn_CAT_sum_f9_to11
dtype: float64
- name: time_MeanAttn_CAT_sum_to11
dtype: float64
- name: time_MeanAttn_ro
dtype: float64
- name: time_MeanAttn_ro_str25
dtype: float64
- name: time_MeanAttn_ro_str50
dtype: float64
- name: time_MeanAttn_ro_str85
dtype: float64
- name: time_MeanAttn_ro_str95
dtype: float64
- name: time_MeanAttn_sum
dtype: float64
- name: time_MeanAttn_sum_f10
dtype: float64
- name: time_MeanAttn_sum_f10_to11
dtype: float64
- name: time_MeanAttn_sum_f11
dtype: float64
- name: time_MeanAttn_sum_f6
dtype: float64
- name: time_MeanAttn_sum_f6_to11
dtype: float64
- name: time_MeanAttn_sum_f7
dtype: float64
- name: time_MeanAttn_sum_f7_to11
dtype: float64
- name: time_MeanAttn_sum_f8
dtype: float64
- name: time_MeanAttn_sum_f8_to11
dtype: float64
- name: time_MeanAttn_sum_f9
dtype: float64
- name: time_MeanAttn_sum_f9_to11
dtype: float64
- name: time_MeanAttn_sum_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str25
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str50
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str85
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str95
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f10
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f10_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f6
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f6_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f7
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f7_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f8
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f8_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f9
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_f9_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_CAT_sum_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_ro
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_ro_str25
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_ro_str50
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_ro_str85
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_ro_str95
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_to1_ro
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str25
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str50
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str85
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str95
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f10
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f10_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f6
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f6_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f7
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f7_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f8
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f8_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f9
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_f9_to11
dtype: float64
- name: time_MeanReLUAttnGrad_MeanAttn_sum_to11
dtype: float64
- name: time_taken
dtype: float64
- name: time_transform_AttnGrad
dtype: float64
- name: time_transform_AttnWHeadGrad
dtype: float64
- name: time_transform_CAT
dtype: float64
- name: time_transform_CAT_AttnFrom
dtype: float64
- name: time_transform_MeanAttn
dtype: float64
- name: time_transform_MeanAttnGrad
dtype: float64
- name: time_transform_MeanAttnGrad_MeanAttn
dtype: float64
- name: time_transform_MeanAttn_CAT
dtype: float64
splits:
- name: train
num_bytes: 1255794036
num_examples: 21
download_size: 1523243218
dataset_size: 1255794036
---
# Dataset Card for "hf_datasets_bug1"
[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
提供机构:
NightMachinery
原始信息汇总
数据集特征概述
基本特征
- id (dtype: int64)
注意力平均值特征
- blocks__0__MeanAttn 至 blocks__11__MeanAttn (dtype: float32)
- MeanAttn_ro 至 MeanAttn_ro_str95 (dtype: float32)
注意力梯度特征
- blocks__0__AttnGrad 至 blocks__11__AttnGrad (dtype: float32)
平均注意力梯度特征
- blocks__0__MeanAttnGrad 至 blocks__11__MeanAttnGrad (dtype: float32)
平均ReLU注意力梯度特征
- blocks__0__MeanReLUAttnGrad 至 blocks__11__MeanReLUAttnGrad (dtype: float32)
混合特征
- blocks__0__MeanAttnGrad_MeanAttn 至 blocks__11__MeanReLUAttnGrad_MeanAttn (dtype: float32)
- MeanAttnGrad_MeanAttn_ro 至 MeanReLUAttnGrad_MeanAttn_relu_to1_ro_str95 (dtype: float32)
CAT特征
- blocks__0__CAT 至 blocks__11__CAT (dtype: float32)
- blocks__0__CAT_AttnFrom 至 blocks__11__CAT_AttnFrom (dtype: float32)
- blocks__0__MeanAttn_CAT 至 blocks__11__MeanReLUAttnGrad_MeanAttn_CAT (dtype: float32)
- MeanAttn_CAT_ro 至 MeanReLUAttnGrad_MeanAttn_CAT_relu_to1_ro_str95 (dtype: float32)
总和特征
- MeanAttn_sum 至 MeanReLUAttnGrad_MeanAttn_sum_f11 (dtype: float32)
- CAT_sum 至 CAT_AttnFrom_sum_f11 (dtype: float32)
- MeanAttn_CAT_sum 至 MeanAttn_CAT_sum_f10_to11 (dtype: float32)
以上特征涵盖了数据集中的主要数据类型和结构,为分析和处理提供了基础信息。



