Update README.md

README.md

@@ -52,13 +52,12 @@ language:
 pipeline_tag: text-generation
 ---
 
-# <span style="color:red"><b>WARNING:</b> Intermediary checkpoint at global step 91100. This checkpoint is not a fully trained model. Evaluations of intermediary checkpoints and the final model will be added when conducted (see below).</span>
 
 # <p>BLOOM LM<br/> _BigScience Large Open-science Open-access Multilingual Language Model_ <br/>Model Card</p>
 <img src="https://assets.website-files.com/6139f3cdcbbff3a68486761d/613cd8997b270da063e230c5_Tekengebied%201-p-500.png" alt="BigScience Logo" width="200"/>
 
 
-Version 1.3 / 3.July.2022 - Checkpoint: **Global step 91100**
+Version 1.3 / 3.July.2022 - Checkpoint: **Global step 95000** - Number of seen tokens: **398B seen tokens**
 
 # Table of Contents
 1. [Model Details](#model-details)
@@ -140,6 +139,7 @@ Please see [the BLOOM training README](https://github.com/bigscience-workshop/bi
     * Sequence length of 2048 tokens used (see [BLOOM tokenizer](https://huggingface.co/bigscience/tokenizer), [tokenizer description](#tokenization))
 
 **Objective Function:** Cross Entropy with mean reduction (see [API documentation](https://pytorch.org/docs/stable/generated/torch.nn.CrossEntropyLoss.html#torch.nn.CrossEntropyLoss)).
+
     
 ### Compute infrastructure
 Jean Zay Public Supercomputer, provided by the French government (see [announcement](https://www.enseignementsup-recherche.gouv.fr/fr/signature-du-marche-d-acquisition-de-l-un-des-supercalculateurs-les-plus-puissants-d-europe-46733)).
@@ -372,6 +372,11 @@ Intentionally using the model for harm, violating [human rights](#human-rights),
 
 -   Generating content without attribution to the model, as specified in the [RAIL License, Use Restrictions](https://huggingface.co/spaces/bigscience/license)
 
+## Intermediate checkpoints
+
+For academic (or any) usage, we published the intermediate checkpoints, corresponding to the model state at each 5000 steps. Please follow [this link](https://huggingface.co/bigscience/bloom-176-intermediate) to get these checkpoints.
+
+
 ## Intended Users
 
 ### Direct Users
@@ -408,6 +413,7 @@ Intentionally using the model for harm, violating [human rights](#human-rights),
 
 ---
 
+
 # Risks and Limitations
 *This section identifies foreseeable harms and misunderstandings.*