evaluate 0 0 0

harpreetsahota/IFEval_Experiments

published by @harpreetsahota

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
473
Last 30 days
13
Likes
1
Size
648 KB
Created on HF
2023-12-12
Age
2 years ago

Tags

size_categories:n<1Kformat:parquetmodality:tabularmodality:textlibrary:datasetslibrary:pandaslibrary:mlcroissantlibrary:polarsregion:us

README current version from Hugging Face


dataset_info:
features:

  • name: key
    dtype: int64
  • name: prompt
    dtype: string
  • name: instruction_id_list
    sequence: string
  • name: kwargs
    dtype: string
  • name: DeciLM_baseline_response
    dtype: string
  • name: DeciLM_baseline_time
    dtype: float64
  • name: Mistral_baseline_response
    dtype: string
  • name: Mistral_baseline_time
    dtype: float64
  • name: DeciLM_greedy_search_response
    dtype: string
  • name: DeciLM_greedy_search_time
    dtype: float64
  • name: Mistral_greedy_search_response
    dtype: string
  • name: Mistral_greedy_search_time
    dtype: float64
  • name: DeciLM_multinomial_sampling_response
    dtype: string
  • name: DeciLM_multinomial_sampling_time
    dtype: float64
  • name: Mistral_multinomial_sampling_response
    dtype: string
  • name: Mistral_multinomial_sampling_time
    dtype: float64
  • name: DeciLM_beam_search_response
    dtype: string
  • name: DeciLM_beam_search_time
    dtype: float64
  • name: Mistral_beam_search_response
    dtype: string
  • name: Mistral_beam_search_time
    dtype: float64
  • name: DeciLM_beam_search_multinomial_response
    dtype: string
  • name: DeciLM_beam_search_multinomial_time
    dtype: float64
  • name: Mistral_beam_search_multinomial_response
    dtype: string
  • name: Mistral_beam_search_multinomial_time
    dtype: float64
  • name: DeciLM_contrastive_search_response
    dtype: string
  • name: DeciLM_contrastive_search_time
    dtype: float64
  • name: Mistral_contrastive_search_response
    dtype: string
  • name: Mistral_contrastive_search_time
    dtype: float64
    splits:
  • name: train
    num_bytes: 1332493
    num_examples: 100
    download_size: 659731
    dataset_size: 1332493
    configs:
  • config_name: default
    data_files:
    • split: train
      path: data/train-*

README history 5 revisions

Every night we snapshot the README of every dataset in the catalog. When the SHA changes we archive the new version and diff it against the last. This is the evolving thought record of the alignment-data field: what the author decided to say about the corpus, and how that framing shifted over time.

  1. 2023-12-17Upload dataset1853fcd1.8 KB
    Loading...
  2. 2023-12-15Upload dataset5faa51d2.8 KB
    Loading...
  3. 2023-12-12Upload dataset5f0ea8a1.7 KB
    Loading...
  4. 2023-12-12Upload dataset27f91ca1.7 KB
    Loading...
  5. 2023-12-12initial commit152fee30 B
    Loading...

Used by abliterated models none yet

No abliterated models in our catalog have declared this dataset in their YAML metadata yet. This can mean the dataset is used but not declared, is used outside abliteration workflows, or is new. The link will populate automatically as models are indexed.

Community discussions 1 thread · 1 open · 1 comments

Threads posted by the community on Hugging Face: bug reports, questions about usage, pull requests on the README, license clarifications. Click any thread to expand and read the full conversation inline. Newest first.

  1. 2023-12-20[bot] Conversion to Parquetopen1 💬#1
    Loading...

Related datasets same stage · evaluate

Read further

Repository

Download the dataset

0

Copy any of these snippets into your notebook or terminal. All three fetch directly from Hugging Face using your own credentials.

Recommended: datasets library (Python)
from datasets import load_dataset
ds = load_dataset("harpreetsahota/IFEval_Experiments")
Raw snapshot (Python)
from huggingface_hub import snapshot_download
snapshot_download(repo_id="harpreetsahota/IFEval_Experiments", repo_type="dataset")
git clone (requires git-lfs)
git clone https://huggingface.co/datasets/harpreetsahota/IFEval_Experiments
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration