{
  "attention_probs_dropout_prob": 0.1,
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "intermediate_size": 2560,
  "max_position_embeddings": 512,
  "position_bucket_size": 32,
  "num_attention_heads": 12,
  "num_hidden_layers": 12,
  "vocab_size": 16384,
  "layer_norm_eps": 1e-05
}

Tokenizer file: tokenizer_eng.json

Quick Usage

python

from transformers import AutoTokenizer, AutoModelForMaskedLM
model_id = 'haznitrama/babybabellm-multi_gpu-gpt_bert-eng-main-causal'
tok = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForMaskedLM.from_pretrained(model_id, trust_remote_code=True)
out = model(**tok('Hello world', return_tensors='pt'))

Causal LM Wrapper

This repo includes a lightweight GPTBertForCausalLM wrapper. Generation example:

python

from transformers import AutoTokenizer, AutoModelForCausalLM
mid='haznitrama/babybabellm-multi_gpu-gpt_bert-eng-main-causal'
tok=AutoTokenizer.from_pretrained(mid)
model=AutoModelForCausalLM.from_pretrained(mid, trust_remote_code=True)
print(tok.decode(model.generate(**tok('Hello', return_tensors='pt'), max_new_tokens=20)[0], skip_special_tokens=True))

Notes

Converted on 2025-09-27T15:21:53.977598+00:00
Weights are the exact trained parameters; no new layers were initialized.
Requires trust_remote_code=True due to custom architecture.