push 774M model

DEVELOPERS.md

@@ -27,8 +27,9 @@ pip3 install -r requirements.txt
 
 Download the model data
 ```
-python3 download_model.py 117M
-python3 download_model.py 345M
+python3 download_model.py 124M
+python3 download_model.py 355M
+python3 download_model.py 774M
 ```
 
 ## Docker Installation

Dockerfile.cpu

@@ -5,5 +5,6 @@ RUN mkdir /gpt-2
 WORKDIR /gpt-2
 ADD . /gpt-2
 RUN pip3 install -r requirements.txt
-RUN python3 download_model.py 117M
-RUN python3 download_model.py 345M
+RUN python3 download_model.py 124M
+RUN python3 download_model.py 355M
+RUN python3 download_model.py 774M

Dockerfile.gpu

@@ -14,5 +14,6 @@ RUN mkdir /gpt-2
 WORKDIR /gpt-2
 ADD . /gpt-2
 RUN pip3 install -r requirements.txt
-RUN python3 download_model.py 117M
-RUN python3 download_model.py 345M
+RUN python3 download_model.py 124M
+RUN python3 download_model.py 355M
+RUN python3 download_model.py 774M

README.md

@@ -4,9 +4,11 @@
 
 Code from the paper ["Language Models are Unsupervised Multitask Learners"](https://d4mucfpksywv.cloudfront.net/better-language-models/language-models.pdf).
 
-We have currently released small (117M parameter) and medium (345M parameter) versions of GPT-2.  While we have not released the larger models, we have [released a dataset](https://github.com/openai/gpt-2-output-dataset) for researchers to study their behaviors.
+We have currently released small (124M parameter), medium (355M parameter), and large (774M parameter) versions of GPT-2<sup>*</sup>, with only the full model as of yet unreleased.  We have also [released a dataset](https://github.com/openai/gpt-2-output-dataset) for researchers to study their behaviors.
 
-See more details in our [blog post](https://blog.openai.com/better-language-models/).
+You can read about GPT-2 and release decisions in our [original blog post](https://blog.openai.com/better-language-models/) and [6 month follow-up post](https://openai.com/blog/gpt-2-6-month-follow-up/).
+
+<sup>*</sup> *Note that our original parameter counts were wrong due to an error (in our previous blog posts and paper).  Thus you may have seen small referred to as 117M and medium referred to as 345M.*
 
 ## Usage
 

download_model.py

@@ -4,7 +4,7 @@
 from tqdm import tqdm
 
 if len(sys.argv) != 2:
-    print('You must enter the model name as a parameter, e.g.: download_model.py 117M')
+    print('You must enter the model name as a parameter, e.g.: download_model.py 124M')
     sys.exit(1)
 
 model = sys.argv[1]

src/generate_unconditional_samples.py

@@ -9,7 +9,7 @@
 import model, sample, encoder
 
 def sample_model(
-    model_name='117M',
+    model_name='124M',
     seed=None,
     nsamples=0,
     batch_size=1,
@@ -20,7 +20,7 @@ def sample_model(
 ):
     """
     Run the sample_model
-    :model_name=117M : String, which model to use
+    :model_name=124M : String, which model to use
     :seed=None : Integer seed for random number generators, fix seed to
      reproduce results
     :nsamples=0 : Number of samples to return, if 0, continues to

src/interactive_conditional_samples.py

@@ -9,18 +9,18 @@
 import model, sample, encoder
 
 def interact_model(
-    model_name='117M',
+    model_name='124M',
     seed=None,
     nsamples=1,
     batch_size=1,
     length=None,
     temperature=1,
     top_k=0,
-    models_dir='models',    
+    models_dir='models',
 ):
     """
     Interactively run the model
-    :model_name=117M : String, which model to use
+    :model_name=124M : String, which model to use
     :seed=None : Integer seed for random number generators, fix seed to reproduce
      results
     :nsamples=1 : Number of samples to return total
@@ -36,7 +36,7 @@ def interact_model(
      while 40 means 40 words are considered at each step. 0 (default) is a
      special setting meaning no restrictions. 40 generally is a good value.
      :models_dir : path to parent folder containing model subfolders
-     (i.e. contains the <model_name> folder)     
+     (i.e. contains the <model_name> folder)
     """
     models_dir = os.path.expanduser(os.path.expandvars(models_dir))
     if batch_size is None: