Firefly(流萤): 中文对话式大语言模型(全量微调+QLoRA),支持微调Llma2、Llama、Qwen、Baichuan、ChatGLM2、InternLM、Ziya、Bloom等大模型 - 修复QWenTokenizer只有eod_id的问题,兼容所有tokenizer · googx/Firefly@67dd449
"tokenizer.eos_token_id # use tokenizer.eod_id instead" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "tokenizer.pad_token_id " ] }, { "cell_type": "code", "execution_count": 11, ...
eod_mask_loss ... False eval_interval ... 1000 eval_iters ... 100 evidence_data_path ... None exit_duration_in_mins ... None exit_interval ... None exit_on_missing
We read every piece of feedback, and take your input very seriously. Include my email address so I can be contacted Cancel Submit feedback Saved searches Use saved searches to filter your results more quickly Cancel Create saved search Sign in Sign up Reseting focus {...
"tokenizer.encode(\"print('<|endoftext|>')\", allowed_special=set(), disallowed_special='all') + [tokenizer.eod_id]\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { ...
Ġbl ood in ess l ing Ġle g ov ed om a Ġe l Ġsk in Ġcons id Ġqu est Ġocc ur Ġf ather Ġm om Ġm us and s Ġhand s Ġsm all p s t en Ġto ok ar k Ġg et Ġy ears Ġsu dden Ġk e Ġb or Ġm ight m en Ġp ur Ġon ce ) . ...