Hugging Face's logo

silver
/
chatglm-6b-slim

Transformers
PyTorch
Chinese
English
chatglm
glm
chatglm-slim
custom_code
Model card Files
xet
 Community
5
chatglm-6b-slim

Commit History

  1. Merge remote-tracking branch 'thu/main'

    silver
    0d1d6a6

  2. Fix position ids expand

    duzx16
    f82b180

  1. Fix generate

    duzx16
    fb23542

  1. Fix attention mask for prefix prompt

    duzx16
    08bc851

  2. No padding for chat function

    duzx16
    4b7ffbf

  3. Fix attention_mask and position_ids

    duzx16
    373fd6b

  4. Fix encode method

    duzx16
    e22cddf

  5. Fix batch input

    duzx16
    e1494f2

  6. Implement batch generation

    duzx16
    cc96a22

  1. Fix position id for training

    duzx16
    11c270c

  2. fix GLM6BBlock name typo (#20)⁠

    zxdu20 +1: richbrain
    9c7416d

  3. Add support for loading quantized model

    duzx16
    2e1be30

  1. Use dynamic dtype for prompts

    duzx16
    c949d03

  2. Fix backward for quantization

    duzx16
    0cfae21

  3. Implement gradient checkpointing

    duzx16
    aea6cef

  4. Fix bugs

    duzx16
    0564795

  1. Add pad_token_id in config.json⁠

    duzx16
    2200e2b

  2. Merge remote-tracking branch 'thu/main'

    silver
    a6d4a44

  3. Change padding side

    duzx16
    db22499

  4. Set ignore_index for CrossEntropyLoss

    duzx16
    5c64357

  5. Support batch training

    duzx16
    8127ab6

  6. Merge branch 'main' into dev_pt⁠

    duzx16
    fbda120

  7. Add p-tuning v2

    duzx16
    812f43f

  1. Fix context length in get_position_ids

    duzx16
    096f3de

  1. update readme

    silver
    a748e08

  2. Merge remote-tracking branch 'thu/main'

    silver
    fb1c9f0

  1. Close CPU fusion on Mac

    duzx16
    4a9b711

  2. Merge remote-tracking branch 'thu/main'

    silver
    a23d6a1

  1. Fix Chinese punctuation

    duzx16
    d2bbc82

  1. Add English

    duzx16
    2449bdc

  1. merge from ChatGLM-6B

    silver
    b8b422f

  1. Fix typo in tokenization_chatglm.py

    songxxzp
    1b54948

  1. fix import issue

    silver
    0b2264f

  2. remove image tokens from chatglm-6b

    silver
    99c7042

  1. Remove hardcode bos_token_id

    duzx16
    8c4ad86

  2. Add support for streaming output

    duzx16
    258e6db

  3. Remove hardcode bos_token_id

    duzx16
    2460dc2

  4. Add support for streaming output

    duzx16
    42095d4

  1. fix import issue

    silver
    ec054ef

  2. update README.md

    silver
    42cd818

  3. remove image tokens from chatglm-6b

    silver
    82834c6

  1. Fix overflow in FP16

    duzx16
    220f772

  1. Set is_parallelizable to False

    duzx16
    f9f74fd

  2. Remove image tokens when decoding

    duzx16
    8492687

  3. Revert "Update quantization"⁠

    duzx16
    0d8b08d

  4. Update quantization

    duzx16
    0101afd

  5. Add logit processor for NaN or Inf scores

    duzx16
    c3dece3

  1. Fix default history argument

    duzx16
    9d1509a

  2. Merge branch 'main' of https://huggingface.co/THUDM/chatglm-6b

    duzx16
    bcb053b

  3. Add support for float32

    duzx16
    d4832e8