htlou commited on
Commit
a817e85
·
verified ·
1 Parent(s): 02ef062

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +1 -0
  2. arguments.yaml +55 -0
  3. environ.txt +225 -0
  4. script.sh +49 -0
  5. slice_12208/added_tokens.json +4 -0
  6. slice_12208/chat_template.json +3 -0
  7. slice_12208/config.json +85 -0
  8. slice_12208/preprocessor_config.json +52 -0
  9. slice_12208/processor_config.json +7 -0
  10. slice_12208/pytorch_model.bin +3 -0
  11. slice_12208/special_tokens_map.json +31 -0
  12. slice_12208/tokenizer.json +0 -0
  13. slice_12208/tokenizer.model +3 -0
  14. slice_12208/tokenizer_config.json +69 -0
  15. slice_24416/added_tokens.json +4 -0
  16. slice_24416/chat_template.json +3 -0
  17. slice_24416/config.json +85 -0
  18. slice_24416/preprocessor_config.json +52 -0
  19. slice_24416/processor_config.json +7 -0
  20. slice_24416/pytorch_model.bin +3 -0
  21. slice_24416/special_tokens_map.json +31 -0
  22. slice_24416/tokenizer.json +0 -0
  23. slice_24416/tokenizer.model +3 -0
  24. slice_24416/tokenizer_config.json +69 -0
  25. slice_36624/added_tokens.json +4 -0
  26. slice_36624/chat_template.json +3 -0
  27. slice_36624/config.json +85 -0
  28. slice_36624/preprocessor_config.json +52 -0
  29. slice_36624/processor_config.json +7 -0
  30. slice_36624/pytorch_model.bin +3 -0
  31. slice_36624/special_tokens_map.json +31 -0
  32. slice_36624/tokenizer.json +0 -0
  33. slice_36624/tokenizer.model +3 -0
  34. slice_36624/tokenizer_config.json +69 -0
  35. slice_48832/added_tokens.json +4 -0
  36. slice_48832/chat_template.json +3 -0
  37. slice_48832/config.json +85 -0
  38. slice_48832/preprocessor_config.json +52 -0
  39. slice_48832/processor_config.json +7 -0
  40. slice_48832/pytorch_model.bin +3 -0
  41. slice_48832/special_tokens_map.json +31 -0
  42. slice_48832/tokenizer.json +0 -0
  43. slice_48832/tokenizer.model +3 -0
  44. slice_48832/tokenizer_config.json +69 -0
  45. slice_61040/added_tokens.json +4 -0
  46. slice_61040/chat_template.json +3 -0
  47. slice_61040/config.json +85 -0
  48. slice_61040/preprocessor_config.json +52 -0
  49. slice_61040/processor_config.json +7 -0
  50. slice_61040/pytorch_model.bin +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ wandb/run-20250330_124344-uzpzatp0/run-uzpzatp0.wandb filter=lfs diff=lfs merge=lfs -text
arguments.yaml ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ data_cfgs:
2
+ eval_data_files: {}
3
+ eval_datasets: {}
4
+ eval_name: {}
5
+ eval_optional_args: []
6
+ eval_size: {}
7
+ eval_split: {}
8
+ eval_subset: {}
9
+ eval_template: {}
10
+ load_multi_datasets: false
11
+ train_data_files: {}
12
+ train_datasets: /aifs4su/yaodong/hantao/datasets/MMInstruct-GPT4V_mistral-7b_cosi_cut/merged/top1-100_valid
13
+ train_name: text-image-to-text
14
+ train_optional_args: []
15
+ train_size: {}
16
+ train_split: train
17
+ train_template: MM_TI2T_LLAVA
18
+ logger_cfgs:
19
+ cache_dir: {}
20
+ log_project: align-anything
21
+ log_run_name: sft
22
+ log_type: wandb
23
+ output_dir: ../outputs/LLAVA_7B_cosi/top1-100_valid
24
+ save_total_limit: 6
25
+ model_cfgs:
26
+ model_max_length: 4096
27
+ model_name_or_path: /aifs4su/yaodong/hantao/models/llava-v1.6-mistral-7b-hf
28
+ trust_remote_code: true
29
+ special_tokens: {}
30
+ train_cfgs:
31
+ adam_betas:
32
+ - 0.9
33
+ - 0.95
34
+ adam_epsilon: 1.0e-08
35
+ bf16: true
36
+ ds_cfgs: ds_z3_config.json
37
+ epochs: 3
38
+ eval_interval: 10
39
+ eval_strategy: epoch
40
+ fp16: false
41
+ freeze_language_model: false
42
+ freeze_mm_proj: false
43
+ freeze_vision_tower: true
44
+ gradient_accumulation_steps: 16
45
+ gradient_checkpointing: true
46
+ learning_rate: 2.0e-05
47
+ load_checkpoint: false
48
+ lr_scheduler_type: cosine
49
+ lr_warmup_ratio: 0.03
50
+ max_grad_norm: 1.0
51
+ per_device_eval_batch_size: 1
52
+ per_device_train_batch_size: 1
53
+ save_checkpoint: false
54
+ seed: 42
55
+ weight_decay: 0.0
environ.txt ADDED
@@ -0,0 +1,225 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ADDR2LINE=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-addr2line
2
+ AR=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-ar
3
+ AS=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-as
4
+ BASH_FUNC__module_raw%%=() { unset _mlshdbg;
5
+ if [ "${MODULES_SILENT_SHELL_DEBUG:-0}" = '1' ]; then
6
+ case "$-" in
7
+ *v*x*)
8
+ set +vx;
9
+ _mlshdbg='vx'
10
+ ;;
11
+ *v*)
12
+ set +v;
13
+ _mlshdbg='v'
14
+ ;;
15
+ *x*)
16
+ set +x;
17
+ _mlshdbg='x'
18
+ ;;
19
+ *)
20
+ _mlshdbg=''
21
+ ;;
22
+ esac;
23
+ fi;
24
+ unset _mlre _mlIFS;
25
+ if [ -n "${IFS+x}" ]; then
26
+ _mlIFS=$IFS;
27
+ fi;
28
+ IFS=' ';
29
+ for _mlv in ${MODULES_RUN_QUARANTINE:-};
30
+ do
31
+ if [ "${_mlv}" = "${_mlv##*[!A-Za-z0-9_]}" -a "${_mlv}" = "${_mlv#[0-9]}" ]; then
32
+ if [ -n "`eval 'echo ${'$_mlv'+x}'`" ]; then
33
+ _mlre="${_mlre:-}${_mlv}_modquar='`eval 'echo ${'$_mlv'}'`' ";
34
+ fi;
35
+ _mlrv="MODULES_RUNENV_${_mlv}";
36
+ _mlre="${_mlre:-}${_mlv}='`eval 'echo ${'$_mlrv':-}'`' ";
37
+ fi;
38
+ done;
39
+ if [ -n "${_mlre:-}" ]; then
40
+ eval `eval ${_mlre} /usr/bin/tclsh /cm/local/apps/environment-modules/4.5.3/libexec/modulecmd.tcl bash '"$@"'`;
41
+ else
42
+ eval `/usr/bin/tclsh /cm/local/apps/environment-modules/4.5.3/libexec/modulecmd.tcl bash "$@"`;
43
+ fi;
44
+ _mlstatus=$?;
45
+ if [ -n "${_mlIFS+x}" ]; then
46
+ IFS=$_mlIFS;
47
+ else
48
+ unset IFS;
49
+ fi;
50
+ unset _mlre _mlv _mlrv _mlIFS;
51
+ if [ -n "${_mlshdbg:-}" ]; then
52
+ set -$_mlshdbg;
53
+ fi;
54
+ unset _mlshdbg;
55
+ return $_mlstatus
56
+ }
57
+ BASH_FUNC_ml%%=() { module ml "$@"
58
+ }
59
+ BASH_FUNC_module%%=() { _module_raw "$@" 2>&1
60
+ }
61
+ BASH_FUNC_switchml%%=() { typeset swfound=1;
62
+ if [ "${MODULES_USE_COMPAT_VERSION:-0}" = '1' ]; then
63
+ typeset swname='main';
64
+ if [ -e /cm/local/apps/environment-modules/4.5.3/libexec/modulecmd.tcl ]; then
65
+ typeset swfound=0;
66
+ unset MODULES_USE_COMPAT_VERSION;
67
+ fi;
68
+ else
69
+ typeset swname='compatibility';
70
+ if [ -e /cm/local/apps/environment-modules/4.5.3/libexec/modulecmd-compat ]; then
71
+ typeset swfound=0;
72
+ MODULES_USE_COMPAT_VERSION=1;
73
+ export MODULES_USE_COMPAT_VERSION;
74
+ fi;
75
+ fi;
76
+ if [ $swfound -eq 0 ]; then
77
+ echo "Switching to Modules $swname version";
78
+ source /cm/local/apps/environment-modules/4.5.3/init/bash;
79
+ else
80
+ echo "Cannot switch to Modules $swname version, command not found";
81
+ return 1;
82
+ fi
83
+ }
84
+ BROWSER=/home/yangyaodong/.vscode-server/cli/servers/Stable-e54c774e0add60467559eb0d1e229c6452cf8447/server/bin/helpers/browser.sh
85
+ BUILD=x86_64-conda-linux-gnu
86
+ BUNDLED_DEBUGPY_PATH=/home/yangyaodong/.vscode-server/extensions/ms-python.debugpy-2025.0.1-linux-x64/bundled/libs/debugpy
87
+ CC=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-cc
88
+ CC_FOR_BUILD=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-cc
89
+ CFLAGS=-march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -ffunction-sections -pipe -isystem /aifs4su/yaodong/miniconda3/envs/hantao_llama/include -I/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/include -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/lib -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/lib/stubs
90
+ CMAKE_ARGS=-DCMAKE_AR=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-ar -DCMAKE_RANLIB=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-ranlib -DCMAKE_LINKER=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-ld -DCMAKE_STRIP=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-strip -DCMAKE_BUILD_TYPE=Release
91
+ CMAKE_PREFIX_PATH=/aifs4su/yaodong/miniconda3/envs/hantao_llama:/aifs4su/yaodong/miniconda3/envs/hantao_llama/x86_64-conda-linux-gnu/sysroot/usr
92
+ CMD_WLM_CLUSTER_NAME=slurm
93
+ COLORTERM=truecolor
94
+ CONDA_BUILD_SYSROOT=/aifs4su/yaodong/miniconda3/envs/hantao_llama/x86_64-conda-linux-gnu/sysroot
95
+ CONDA_DEFAULT_ENV=hantao_llama
96
+ CONDA_EXE=/aifs4su/yaodong/miniconda3/bin/conda
97
+ CONDA_PREFIX=/aifs4su/yaodong/miniconda3/envs/hantao_llama
98
+ CONDA_PREFIX_1=/aifs4su/yaodong/miniconda3
99
+ CONDA_PROMPT_MODIFIER=(hantao_llama)
100
+ CONDA_PYTHON_EXE=/aifs4su/yaodong/miniconda3/bin/python
101
+ CONDA_SHLVL=2
102
+ CPATH=/cm/shared/apps/slurm/current/include
103
+ CPATH_modshare=/cm/shared/apps/slurm/current/include:1
104
+ CPP=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-cpp
105
+ CPPFLAGS=-DNDEBUG -D_FORTIFY_SOURCE=2 -O2 -isystem /aifs4su/yaodong/miniconda3/envs/hantao_llama/include -I/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/include -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/lib -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/lib/stubs
106
+ CROSS_RANK=0
107
+ CROSS_SIZE=1
108
+ CUDA_MODULE_LOADING=LAZY
109
+ CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
110
+ CXX=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-c++
111
+ CXXFILT=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-c++filt
112
+ CXXFLAGS=-fvisibility-inlines-hidden -std=c++17 -fmessage-length=0 -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -ffunction-sections -pipe -isystem /aifs4su/yaodong/miniconda3/envs/hantao_llama/include -I/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/include -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/lib -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/lib/stubs
113
+ CXX_FOR_BUILD=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-c++
114
+ DBUS_SESSION_BUS_ADDRESS=unix:path=/run/user/1028/bus
115
+ DEBUGPY_ADAPTER_ENDPOINTS=/home/yangyaodong/.vscode-server/extensions/ms-python.debugpy-2025.0.1-linux-x64/.noConfigDebugAdapterEndpoints/endpoint-cf2a8fd1c0b5bb2d.txt
116
+ DEBUG_CFLAGS=-march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-all -fno-plt -Og -g -Wall -Wextra -fvar-tracking-assignments -ffunction-sections -pipe -isystem /aifs4su/yaodong/miniconda3/envs/hantao_llama/include
117
+ DEBUG_CPPFLAGS=-D_DEBUG -D_FORTIFY_SOURCE=2 -Og -isystem /aifs4su/yaodong/miniconda3/envs/hantao_llama/include
118
+ DEBUG_CXXFLAGS=-fvisibility-inlines-hidden -std=c++17 -fmessage-length=0 -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-all -fno-plt -Og -g -Wall -Wextra -fvar-tracking-assignments -ffunction-sections -pipe -isystem /aifs4su/yaodong/miniconda3/envs/hantao_llama/include
119
+ ELFEDIT=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-elfedit
120
+ ENABLE_LMOD=0
121
+ GCC=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-gcc
122
+ GCC_AR=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-gcc-ar
123
+ GCC_NM=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-gcc-nm
124
+ GCC_RANLIB=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-gcc-ranlib
125
+ GIT_ASKPASS=/home/yangyaodong/.vscode-server/cli/servers/Stable-e54c774e0add60467559eb0d1e229c6452cf8447/server/extensions/git/dist/askpass.sh
126
+ GPROF=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-gprof
127
+ GSETTINGS_SCHEMA_DIR=/aifs4su/yaodong/miniconda3/envs/hantao_llama/share/glib-2.0/schemas
128
+ GSETTINGS_SCHEMA_DIR_CONDA_BACKUP=
129
+ GXX=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-g++
130
+ HF_DATASETS_CACHE=/aifs4su/yaodong/.cache/huggingface/datasets
131
+ HF_HOME=/aifs4su/yaodong/.cache/huggingface
132
+ HISTTIMEFORMAT=%y/%m/%d %T
133
+ HOME=/home/yangyaodong
134
+ HOST=x86_64-conda-linux-gnu
135
+ KMP_DUPLICATE_LIB_OK=True
136
+ KMP_INIT_AT_FORK=FALSE
137
+ LANG=C.UTF-8
138
+ LD=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-ld
139
+ LDFLAGS=-Wl,-O2 -Wl,--sort-common -Wl,--as-needed -Wl,-z,relro -Wl,-z,now -Wl,--disable-new-dtags -Wl,--gc-sections -Wl,-rpath,/aifs4su/yaodong/miniconda3/envs/hantao_llama/lib -Wl,-rpath-link,/aifs4su/yaodong/miniconda3/envs/hantao_llama/lib -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/lib -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/lib -L/aifs4su/yaodong/miniconda3/envs/hantao_llama/targets/x86_64-linux/lib/stubs
140
+ LD_GOLD=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-ld.gold
141
+ LD_LIBRARY_PATH=/aifs4su/yaodong/miniconda3/envs/hantao_llama/lib/python3.11/site-packages/cv2/../../lib64:/usr/mpi/gcc/openmpi-4.1.7a1/lib:/cm/shared/apps/slurm/current/lib64/slurm:/cm/shared/apps/slurm/current/lib64
142
+ LD_LIBRARY_PATH_modshare=/cm/shared/apps/slurm/current/lib64:1:/usr/mpi/gcc/openmpi-4.1.7a1/lib:1:/cm/shared/apps/slurm/current/lib64/slurm:1
143
+ LD_RUN_PATH=/usr/mpi/gcc/openmpi-4.1.7a1/lib
144
+ LD_RUN_PATH_modshare=/usr/mpi/gcc/openmpi-4.1.7a1/lib:1
145
+ LESSCLOSE=/usr/bin/lesspipe %s %s
146
+ LESSOPEN=| /usr/bin/lesspipe %s
147
+ LIBRARY_PATH=/cm/shared/apps/slurm/current/lib64/slurm:/cm/shared/apps/slurm/current/lib64
148
+ LIBRARY_PATH_modshare=/cm/shared/apps/slurm/current/lib64:1:/cm/shared/apps/slurm/current/lib64/slurm:1
149
+ LOADEDMODULES=slurm/slurm/23.02.6:gcc/64/4.1.7a1
150
+ LOADEDMODULES_modshare=slurm/slurm/23.02.6:1:gcc/64/4.1.7a1:1
151
+ LOCAL_RANK=0
152
+ LOCAL_SIZE=8
153
+ LOGLEVEL=WARNING
154
+ LOGNAME=yangyaodong
155
+ LS_COLORS=rs=0:di=01;34:ln=01;36:mh=00:pi=40;33:so=01;35:do=01;35:bd=40;33;01:cd=40;33;01:or=40;31;01:mi=00:su=37;41:sg=30;43:ca=30;41:tw=30;42:ow=34;42:st=37;44:ex=01;32:*.tar=01;31:*.tgz=01;31:*.arc=01;31:*.arj=01;31:*.taz=01;31:*.lha=01;31:*.lz4=01;31:*.lzh=01;31:*.lzma=01;31:*.tlz=01;31:*.txz=01;31:*.tzo=01;31:*.t7z=01;31:*.zip=01;31:*.z=01;31:*.dz=01;31:*.gz=01;31:*.lrz=01;31:*.lz=01;31:*.lzo=01;31:*.xz=01;31:*.zst=01;31:*.tzst=01;31:*.bz2=01;31:*.bz=01;31:*.tbz=01;31:*.tbz2=01;31:*.tz=01;31:*.deb=01;31:*.rpm=01;31:*.jar=01;31:*.war=01;31:*.ear=01;31:*.sar=01;31:*.rar=01;31:*.alz=01;31:*.ace=01;31:*.zoo=01;31:*.cpio=01;31:*.7z=01;31:*.rz=01;31:*.cab=01;31:*.wim=01;31:*.swm=01;31:*.dwm=01;31:*.esd=01;31:*.jpg=01;35:*.jpeg=01;35:*.mjpg=01;35:*.mjpeg=01;35:*.gif=01;35:*.bmp=01;35:*.pbm=01;35:*.pgm=01;35:*.ppm=01;35:*.tga=01;35:*.xbm=01;35:*.xpm=01;35:*.tif=01;35:*.tiff=01;35:*.png=01;35:*.svg=01;35:*.svgz=01;35:*.mng=01;35:*.pcx=01;35:*.mov=01;35:*.mpg=01;35:*.mpeg=01;35:*.m2v=01;35:*.mkv=01;35:*.webm=01;35:*.webp=01;35:*.ogm=01;35:*.mp4=01;35:*.m4v=01;35:*.mp4v=01;35:*.vob=01;35:*.qt=01;35:*.nuv=01;35:*.wmv=01;35:*.asf=01;35:*.rm=01;35:*.rmvb=01;35:*.flc=01;35:*.avi=01;35:*.fli=01;35:*.flv=01;35:*.gl=01;35:*.dl=01;35:*.xcf=01;35:*.xwd=01;35:*.yuv=01;35:*.cgm=01;35:*.emf=01;35:*.ogv=01;35:*.ogx=01;35:*.aac=00;36:*.au=00;36:*.flac=00;36:*.m4a=00;36:*.mid=00;36:*.midi=00;36:*.mka=00;36:*.mp3=00;36:*.mpc=00;36:*.ogg=00;36:*.ra=00;36:*.wav=00;36:*.oga=00;36:*.opus=00;36:*.spx=00;36:*.xspf=00;36:
156
+ MANPATH=/usr/mpi/gcc/openmpi-4.1.7a1/share/man:/cm/shared/apps/slurm/current/man:/cm/local/apps/environment-modules/4.5.3/share/man:/usr/local/man:/usr/local/share/man:/usr/share/man:/cm/local/apps/environment-modules/current/share/man:/cm/local/apps/environment-modules/current/share/man
157
+ MANPATH_modshare=/usr/local/share/man:1:/usr/mpi/gcc/openmpi-4.1.7a1/share/man:1:/cm/local/apps/environment-modules/current/share/man:1:/cm/local/apps/environment-modules/4.5.3/share/man:1:/usr/local/man:1:/usr/share/man:1:/cm/shared/apps/slurm/current/man:1
158
+ MASTER_ADDR=127.0.0.1
159
+ MASTER_PORT=50192
160
+ MIG_PARTED_CHECKPOINT_FILE=/var/lib/nvidia-mig-manager/checkpoint.json
161
+ MIG_PARTED_CONFIG_FILE=/etc/nvidia-mig-manager/config.yaml
162
+ MIG_PARTED_HOOKS_FILE=/etc/nvidia-mig-manager/hooks.yaml
163
+ MODULEPATH=/cm/local/modulefiles:/cm/shared/modulefiles
164
+ MODULESHOME=/cm/local/apps/environment-modules/4.5.3
165
+ MODULES_CMD=/cm/local/apps/environment-modules/4.5.3/libexec/modulecmd.tcl
166
+ MODULES_SET_SHELL_STARTUP=0
167
+ MOTD_SHOWN=pam
168
+ MPI_HOME=/usr/mpi/gcc/openmpi-4.1.7a1
169
+ MPI_RUN=/usr/mpi/gcc/openmpi-4.1.7a1/bin/mpirun
170
+ NM=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-nm
171
+ NVCC_PREPEND_FLAGS= -ccbin=/aifs4su/yaodong/miniconda3/bin/x86_64-conda-linux-gnu-c++ -ccbin=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-c++
172
+ NVCC_PREPEND_FLAGS_BACKUP= -ccbin=/aifs4su/yaodong/miniconda3/bin/x86_64-conda-linux-gnu-c++
173
+ NVITOP_MONITOR_MODE=colorful
174
+ OBJCOPY=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-objcopy
175
+ OBJDUMP=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-objdump
176
+ OLDPWD=/home/yangyaodong
177
+ PATH=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin:/usr/lpp/mmfs/bin:/usr/local/cuda/bin:/opt/bin:/usr/lpp/mmfs/bin:/cm/shared/apps/slurm/current/sbin:/cm/shared/apps/slurm/current/bin:/usr/local/cuda/bin:/opt/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/sbin:/usr/sbin:/cm/local/apps/environment-modules/4.5.3/bin
178
+ PATH_modshare=/usr/mpi/gcc/openmpi-4.1.7a1/bin:1:/opt/bin/:1:/usr/bin:1:/usr/local/bin:1:/cm/shared/apps/slurm/current/bin:1:/home/yangyaodong/.vscode-server/cli/servers/Stable-e54c774e0add60467559eb0d1e229c6452cf8447/server/bin/remote-cli:1:/cm/shared/apps/slurm/current/sbin:1:/bin:1:/snap/bin:1:/sbin:1:/home/yangyaodong/.vscode-server/data/User/globalStorage/github.copilot-chat/debugCommand:1:/home/yangyaodong/.vscode-server/extensions/ms-python.debugpy-2025.0.1-linux-x64/bundled/scripts/noConfigScripts:1:/usr/sbin:1:/usr/games:1:/cm/local/apps/environment-modules/4.5.3/bin:1:/usr/local/sbin:1:/usr/lpp/mmfs/bin:1:/usr/local/cuda/bin:1:/usr/local/games:1
179
+ PWD=/aifs4su/yaodong/hantao/align-anything/scripts
180
+ PYDEVD_DISABLE_FILE_VALIDATION=1
181
+ PYTHONHASHSEED=42
182
+ PYTHONPATH=/aifs4su/yaodong/hantao/align-anything/scripts
183
+ QT_QPA_FONTDIR=/aifs4su/yaodong/miniconda3/envs/hantao_llama/lib/python3.11/site-packages/cv2/qt/fonts
184
+ QT_QPA_PLATFORM_PLUGIN_PATH=/aifs4su/yaodong/miniconda3/envs/hantao_llama/lib/python3.11/site-packages/cv2/qt/plugins
185
+ RANK=0
186
+ RANLIB=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-ranlib
187
+ READELF=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-readelf
188
+ SHELL=/bin/bash
189
+ SHLVL=4
190
+ SIZE=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-size
191
+ SLURM_CONF=/cm/shared/apps/slurm/var/etc/slurm/slurm.conf
192
+ SSH_CLIENT=10.33.4.51 46666 22
193
+ SSH_CONNECTION=10.33.4.230 40638 10.33.4.213 22
194
+ SSL_CERT_DIR=/usr/lib/ssl/certs
195
+ SSL_CERT_FILE=/usr/lib/ssl/certs/ca-certificates.crt
196
+ STRINGS=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-strings
197
+ STRIP=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/x86_64-conda-linux-gnu-strip
198
+ TERM=screen
199
+ TERM_PROGRAM=tmux
200
+ TERM_PROGRAM_VERSION=3.2a
201
+ TMUX=/tmp/tmux-1028/default,2296743,10
202
+ TMUX_PANE=%25
203
+ USER=yangyaodong
204
+ VSCODE_GIT_ASKPASS_EXTRA_ARGS=
205
+ VSCODE_GIT_ASKPASS_MAIN=/home/yangyaodong/.vscode-server/cli/servers/Stable-e54c774e0add60467559eb0d1e229c6452cf8447/server/extensions/git/dist/askpass-main.js
206
+ VSCODE_GIT_ASKPASS_NODE=/home/yangyaodong/.vscode-server/cli/servers/Stable-e54c774e0add60467559eb0d1e229c6452cf8447/server/node
207
+ VSCODE_GIT_IPC_HANDLE=/run/user/1028/vscode-git-bbbbf321f6.sock
208
+ VSCODE_IPC_HOOK_CLI=/run/user/1028/vscode-ipc-e2edf668-dca9-4331-a6ac-7d4507f653ce.sock
209
+ WANDB_API_KEY=7e2dcc0c310ebcb7cdcafd5e9320d6be55cf1a33
210
+ WANDB_SERVICE=2-1109563-tcp-localhost-43957
211
+ WORLD_SIZE=8
212
+ XDG_DATA_DIRS=/usr/local/share:/usr/share:/var/lib/snapd/desktop
213
+ XDG_RUNTIME_DIR=/run/user/1028
214
+ XDG_SESSION_CLASS=user
215
+ XDG_SESSION_ID=43255
216
+ XDG_SESSION_TYPE=tty
217
+ ZERO_STAGE=3
218
+ _=/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/deepspeed
219
+ _CE_CONDA=
220
+ _CE_M=
221
+ _CONDA_PYTHON_SYSCONFIGDATA_NAME=_sysconfigdata_x86_64_conda_cos7_linux_gnu
222
+ _LMFILES_=/cm/local/modulefiles/slurm/slurm/23.02.6:/cm/local/modulefiles/gcc/64/4.1.7a1
223
+ _LMFILES__modshare=/cm/local/modulefiles/slurm/slurm/23.02.6:1:/cm/local/modulefiles/gcc/64/4.1.7a1:1
224
+ build_alias=x86_64-conda-linux-gnu
225
+ host_alias=x86_64-conda-linux-gnu
script.sh ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env bash
2
+ #
3
+ # Copyright 2025 PKU-Alignment Team. All Rights Reserved.
4
+ #
5
+ # Licensed under the Apache License, Version 2.0 (the "License");
6
+ # you may not use this file except in compliance with the License.
7
+ # You may obtain a copy of the License at
8
+ #
9
+ # http://www.apache.org/licenses/LICENSE-2.0
10
+ #
11
+ # Unless required by applicable law or agreed to in writing, software
12
+ # distributed under the License is distributed on an "AS IS" BASIS,
13
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14
+ # See the License for the specific language governing permissions and
15
+ # limitations under the License.
16
+ # ==============================================================================
17
+
18
+ DATASETS_NAME=("top1-100_valid")
19
+
20
+ MODEL_NAME_OR_PATH="/aifs4su/yaodong/hantao/models/llava-v1.6-mistral-7b-hf" # model path
21
+
22
+ for DATASET_NAME in "${DATASETS_NAME[@]}"; do
23
+ TRAIN_DATASETS="/aifs4su/yaodong/hantao/datasets/MMInstruct-GPT4V_mistral-7b_cosi_cut/merged/${DATASET_NAME}" # dataset path
24
+ TRAIN_TEMPLATE="MM_TI2T_LLAVA" # dataset template
25
+ TRAIN_NAME="text-image-to-text" # dataset name
26
+ TRAIN_SPLIT="train" # split the dataset
27
+
28
+ OUTPUT_DIR="../outputs/LLAVA_7B_cosi/${DATASET_NAME}" # output dir
29
+
30
+ # For wandb online logging
31
+ export WANDB_API_KEY="7e2dcc0c310ebcb7cdcafd5e9320d6be55cf1a33"
32
+
33
+ # Source the setup script
34
+ source ./setup.sh
35
+
36
+ # Execute deepspeed command
37
+ deepspeed \
38
+ --master_port ${MASTER_PORT} \
39
+ --module align_anything.trainers.text_image_to_text.sft \
40
+ --model_name_or_path ${MODEL_NAME_OR_PATH} \
41
+ --train_datasets ${TRAIN_DATASETS} \
42
+ --train_template ${TRAIN_TEMPLATE} \
43
+ --train_split ${TRAIN_SPLIT} \
44
+ --train_name ${TRAIN_NAME} \
45
+ --output_dir ${OUTPUT_DIR} \
46
+ --save_total_limit 6 \
47
+ --train_batch_size 16 \
48
+ --epochs 3
49
+ done
slice_12208/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<image>": 32000,
3
+ "<pad>": 32001
4
+ }
slice_12208/chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% for message in messages %}{% if message['role'] != 'system' %}{{ message['role'].upper() + ': '}}{% endif %}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>\n' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ content['text'] + ' '}}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ content['text'] + ' '}}{% endgeneration %}{% endfor %}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'ASSISTANT:' }}{% endif %}"
3
+ }
slice_12208/config.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "LlavaNextForConditionalGeneration"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "ignore_index": -100,
9
+ "image_grid_pinpoints": [
10
+ [
11
+ 336,
12
+ 672
13
+ ],
14
+ [
15
+ 672,
16
+ 336
17
+ ],
18
+ [
19
+ 672,
20
+ 672
21
+ ],
22
+ [
23
+ 1008,
24
+ 336
25
+ ],
26
+ [
27
+ 336,
28
+ 1008
29
+ ]
30
+ ],
31
+ "image_seq_length": 576,
32
+ "image_token_index": 32000,
33
+ "model_type": "llava_next",
34
+ "multimodal_projector_bias": true,
35
+ "pad_token_id": 32001,
36
+ "projector_hidden_act": "gelu",
37
+ "text_config": {
38
+ "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
39
+ "architectures": [
40
+ "MistralForCausalLM"
41
+ ],
42
+ "attention_dropout": 0.0,
43
+ "head_dim": 128,
44
+ "hidden_act": "silu",
45
+ "hidden_size": 4096,
46
+ "initializer_range": 0.02,
47
+ "intermediate_size": 14336,
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "mistral",
50
+ "num_attention_heads": 32,
51
+ "num_hidden_layers": 32,
52
+ "num_key_value_heads": 8,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_theta": 1000000.0,
55
+ "sliding_window": null,
56
+ "torch_dtype": "bfloat16",
57
+ "use_cache": true,
58
+ "vocab_size": 32064
59
+ },
60
+ "tie_word_embeddings": false,
61
+ "torch_dtype": "bfloat16",
62
+ "transformers_version": "4.50.0",
63
+ "use_image_newline_parameter": true,
64
+ "vision_config": {
65
+ "attention_dropout": 0.0,
66
+ "hidden_act": "quick_gelu",
67
+ "hidden_size": 1024,
68
+ "image_size": 336,
69
+ "initializer_factor": 1.0,
70
+ "initializer_range": 0.02,
71
+ "intermediate_size": 4096,
72
+ "layer_norm_eps": 1e-05,
73
+ "model_type": "clip_vision_model",
74
+ "num_attention_heads": 16,
75
+ "num_channels": 3,
76
+ "num_hidden_layers": 24,
77
+ "patch_size": 14,
78
+ "projection_dim": 768,
79
+ "torch_dtype": "bfloat16",
80
+ "vocab_size": 32000
81
+ },
82
+ "vision_feature_layer": -2,
83
+ "vision_feature_select_strategy": "default",
84
+ "vocab_size": 32064
85
+ }
slice_12208/preprocessor_config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "aspect_ratio_setting": "anyres",
3
+ "crop_size": {
4
+ "height": 336,
5
+ "width": 336
6
+ },
7
+ "do_center_crop": true,
8
+ "do_convert_rgb": true,
9
+ "do_normalize": true,
10
+ "do_pad": true,
11
+ "do_rescale": true,
12
+ "do_resize": true,
13
+ "image_grid_pinpoints": [
14
+ [
15
+ 336,
16
+ 672
17
+ ],
18
+ [
19
+ 672,
20
+ 336
21
+ ],
22
+ [
23
+ 672,
24
+ 672
25
+ ],
26
+ [
27
+ 1008,
28
+ 336
29
+ ],
30
+ [
31
+ 336,
32
+ 1008
33
+ ]
34
+ ],
35
+ "image_mean": [
36
+ 0.48145466,
37
+ 0.4578275,
38
+ 0.40821073
39
+ ],
40
+ "image_processor_type": "LlavaNextImageProcessor",
41
+ "image_std": [
42
+ 0.26862954,
43
+ 0.26130258,
44
+ 0.27577711
45
+ ],
46
+ "processor_class": "LlavaNextProcessor",
47
+ "resample": 3,
48
+ "rescale_factor": 0.00392156862745098,
49
+ "size": {
50
+ "shortest_edge": 336
51
+ }
52
+ }
slice_12208/processor_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_token": "<image>",
3
+ "num_additional_image_tokens": 1,
4
+ "patch_size": 14,
5
+ "processor_class": "LlavaNextProcessor",
6
+ "vision_feature_select_strategy": "default"
7
+ }
slice_12208/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e3188f9ab268711dd1ddfd113384c09efcc5afe4771ffddd5a4619b5cdc5221
3
+ size 15133733934
slice_12208/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "image_token": "<image>",
17
+ "pad_token": {
18
+ "content": "<pad>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "unk_token": {
25
+ "content": "<unk>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
slice_12208/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
slice_12208/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
slice_12208/tokenizer_config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "32000": {
31
+ "content": "<image>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "32001": {
39
+ "content": "<pad>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ }
46
+ },
47
+ "additional_special_tokens": [],
48
+ "bos_token": "<s>",
49
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
50
+ "clean_up_tokenization_spaces": false,
51
+ "eos_token": "</s>",
52
+ "extra_special_tokens": {
53
+ "image_token": "<image>"
54
+ },
55
+ "image_token": "<image>",
56
+ "legacy": true,
57
+ "max_length": null,
58
+ "model_max_length": 4096,
59
+ "pad_to_multiple_of": null,
60
+ "pad_token": "<pad>",
61
+ "pad_token_type_id": 0,
62
+ "padding_side": "right",
63
+ "processor_class": "LlavaNextProcessor",
64
+ "sp_model_kwargs": {},
65
+ "spaces_between_special_tokens": false,
66
+ "tokenizer_class": "LlamaTokenizer",
67
+ "unk_token": "<unk>",
68
+ "use_default_system_prompt": false
69
+ }
slice_24416/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<image>": 32000,
3
+ "<pad>": 32001
4
+ }
slice_24416/chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% for message in messages %}{% if message['role'] != 'system' %}{{ message['role'].upper() + ': '}}{% endif %}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>\n' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ content['text'] + ' '}}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ content['text'] + ' '}}{% endgeneration %}{% endfor %}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'ASSISTANT:' }}{% endif %}"
3
+ }
slice_24416/config.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "LlavaNextForConditionalGeneration"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "ignore_index": -100,
9
+ "image_grid_pinpoints": [
10
+ [
11
+ 336,
12
+ 672
13
+ ],
14
+ [
15
+ 672,
16
+ 336
17
+ ],
18
+ [
19
+ 672,
20
+ 672
21
+ ],
22
+ [
23
+ 1008,
24
+ 336
25
+ ],
26
+ [
27
+ 336,
28
+ 1008
29
+ ]
30
+ ],
31
+ "image_seq_length": 576,
32
+ "image_token_index": 32000,
33
+ "model_type": "llava_next",
34
+ "multimodal_projector_bias": true,
35
+ "pad_token_id": 32001,
36
+ "projector_hidden_act": "gelu",
37
+ "text_config": {
38
+ "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
39
+ "architectures": [
40
+ "MistralForCausalLM"
41
+ ],
42
+ "attention_dropout": 0.0,
43
+ "head_dim": 128,
44
+ "hidden_act": "silu",
45
+ "hidden_size": 4096,
46
+ "initializer_range": 0.02,
47
+ "intermediate_size": 14336,
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "mistral",
50
+ "num_attention_heads": 32,
51
+ "num_hidden_layers": 32,
52
+ "num_key_value_heads": 8,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_theta": 1000000.0,
55
+ "sliding_window": null,
56
+ "torch_dtype": "bfloat16",
57
+ "use_cache": true,
58
+ "vocab_size": 32064
59
+ },
60
+ "tie_word_embeddings": false,
61
+ "torch_dtype": "bfloat16",
62
+ "transformers_version": "4.50.0",
63
+ "use_image_newline_parameter": true,
64
+ "vision_config": {
65
+ "attention_dropout": 0.0,
66
+ "hidden_act": "quick_gelu",
67
+ "hidden_size": 1024,
68
+ "image_size": 336,
69
+ "initializer_factor": 1.0,
70
+ "initializer_range": 0.02,
71
+ "intermediate_size": 4096,
72
+ "layer_norm_eps": 1e-05,
73
+ "model_type": "clip_vision_model",
74
+ "num_attention_heads": 16,
75
+ "num_channels": 3,
76
+ "num_hidden_layers": 24,
77
+ "patch_size": 14,
78
+ "projection_dim": 768,
79
+ "torch_dtype": "bfloat16",
80
+ "vocab_size": 32000
81
+ },
82
+ "vision_feature_layer": -2,
83
+ "vision_feature_select_strategy": "default",
84
+ "vocab_size": 32064
85
+ }
slice_24416/preprocessor_config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "aspect_ratio_setting": "anyres",
3
+ "crop_size": {
4
+ "height": 336,
5
+ "width": 336
6
+ },
7
+ "do_center_crop": true,
8
+ "do_convert_rgb": true,
9
+ "do_normalize": true,
10
+ "do_pad": true,
11
+ "do_rescale": true,
12
+ "do_resize": true,
13
+ "image_grid_pinpoints": [
14
+ [
15
+ 336,
16
+ 672
17
+ ],
18
+ [
19
+ 672,
20
+ 336
21
+ ],
22
+ [
23
+ 672,
24
+ 672
25
+ ],
26
+ [
27
+ 1008,
28
+ 336
29
+ ],
30
+ [
31
+ 336,
32
+ 1008
33
+ ]
34
+ ],
35
+ "image_mean": [
36
+ 0.48145466,
37
+ 0.4578275,
38
+ 0.40821073
39
+ ],
40
+ "image_processor_type": "LlavaNextImageProcessor",
41
+ "image_std": [
42
+ 0.26862954,
43
+ 0.26130258,
44
+ 0.27577711
45
+ ],
46
+ "processor_class": "LlavaNextProcessor",
47
+ "resample": 3,
48
+ "rescale_factor": 0.00392156862745098,
49
+ "size": {
50
+ "shortest_edge": 336
51
+ }
52
+ }
slice_24416/processor_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_token": "<image>",
3
+ "num_additional_image_tokens": 1,
4
+ "patch_size": 14,
5
+ "processor_class": "LlavaNextProcessor",
6
+ "vision_feature_select_strategy": "default"
7
+ }
slice_24416/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0e08b154cf652cb73935b1ecb527309676b1c2b41788fdcc464a604134ea392
3
+ size 15133733934
slice_24416/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "image_token": "<image>",
17
+ "pad_token": {
18
+ "content": "<pad>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "unk_token": {
25
+ "content": "<unk>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
slice_24416/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
slice_24416/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
slice_24416/tokenizer_config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "32000": {
31
+ "content": "<image>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "32001": {
39
+ "content": "<pad>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ }
46
+ },
47
+ "additional_special_tokens": [],
48
+ "bos_token": "<s>",
49
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
50
+ "clean_up_tokenization_spaces": false,
51
+ "eos_token": "</s>",
52
+ "extra_special_tokens": {
53
+ "image_token": "<image>"
54
+ },
55
+ "image_token": "<image>",
56
+ "legacy": true,
57
+ "max_length": null,
58
+ "model_max_length": 4096,
59
+ "pad_to_multiple_of": null,
60
+ "pad_token": "<pad>",
61
+ "pad_token_type_id": 0,
62
+ "padding_side": "right",
63
+ "processor_class": "LlavaNextProcessor",
64
+ "sp_model_kwargs": {},
65
+ "spaces_between_special_tokens": false,
66
+ "tokenizer_class": "LlamaTokenizer",
67
+ "unk_token": "<unk>",
68
+ "use_default_system_prompt": false
69
+ }
slice_36624/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<image>": 32000,
3
+ "<pad>": 32001
4
+ }
slice_36624/chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% for message in messages %}{% if message['role'] != 'system' %}{{ message['role'].upper() + ': '}}{% endif %}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>\n' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ content['text'] + ' '}}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ content['text'] + ' '}}{% endgeneration %}{% endfor %}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'ASSISTANT:' }}{% endif %}"
3
+ }
slice_36624/config.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "LlavaNextForConditionalGeneration"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "ignore_index": -100,
9
+ "image_grid_pinpoints": [
10
+ [
11
+ 336,
12
+ 672
13
+ ],
14
+ [
15
+ 672,
16
+ 336
17
+ ],
18
+ [
19
+ 672,
20
+ 672
21
+ ],
22
+ [
23
+ 1008,
24
+ 336
25
+ ],
26
+ [
27
+ 336,
28
+ 1008
29
+ ]
30
+ ],
31
+ "image_seq_length": 576,
32
+ "image_token_index": 32000,
33
+ "model_type": "llava_next",
34
+ "multimodal_projector_bias": true,
35
+ "pad_token_id": 32001,
36
+ "projector_hidden_act": "gelu",
37
+ "text_config": {
38
+ "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
39
+ "architectures": [
40
+ "MistralForCausalLM"
41
+ ],
42
+ "attention_dropout": 0.0,
43
+ "head_dim": 128,
44
+ "hidden_act": "silu",
45
+ "hidden_size": 4096,
46
+ "initializer_range": 0.02,
47
+ "intermediate_size": 14336,
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "mistral",
50
+ "num_attention_heads": 32,
51
+ "num_hidden_layers": 32,
52
+ "num_key_value_heads": 8,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_theta": 1000000.0,
55
+ "sliding_window": null,
56
+ "torch_dtype": "bfloat16",
57
+ "use_cache": true,
58
+ "vocab_size": 32064
59
+ },
60
+ "tie_word_embeddings": false,
61
+ "torch_dtype": "bfloat16",
62
+ "transformers_version": "4.50.0",
63
+ "use_image_newline_parameter": true,
64
+ "vision_config": {
65
+ "attention_dropout": 0.0,
66
+ "hidden_act": "quick_gelu",
67
+ "hidden_size": 1024,
68
+ "image_size": 336,
69
+ "initializer_factor": 1.0,
70
+ "initializer_range": 0.02,
71
+ "intermediate_size": 4096,
72
+ "layer_norm_eps": 1e-05,
73
+ "model_type": "clip_vision_model",
74
+ "num_attention_heads": 16,
75
+ "num_channels": 3,
76
+ "num_hidden_layers": 24,
77
+ "patch_size": 14,
78
+ "projection_dim": 768,
79
+ "torch_dtype": "bfloat16",
80
+ "vocab_size": 32000
81
+ },
82
+ "vision_feature_layer": -2,
83
+ "vision_feature_select_strategy": "default",
84
+ "vocab_size": 32064
85
+ }
slice_36624/preprocessor_config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "aspect_ratio_setting": "anyres",
3
+ "crop_size": {
4
+ "height": 336,
5
+ "width": 336
6
+ },
7
+ "do_center_crop": true,
8
+ "do_convert_rgb": true,
9
+ "do_normalize": true,
10
+ "do_pad": true,
11
+ "do_rescale": true,
12
+ "do_resize": true,
13
+ "image_grid_pinpoints": [
14
+ [
15
+ 336,
16
+ 672
17
+ ],
18
+ [
19
+ 672,
20
+ 336
21
+ ],
22
+ [
23
+ 672,
24
+ 672
25
+ ],
26
+ [
27
+ 1008,
28
+ 336
29
+ ],
30
+ [
31
+ 336,
32
+ 1008
33
+ ]
34
+ ],
35
+ "image_mean": [
36
+ 0.48145466,
37
+ 0.4578275,
38
+ 0.40821073
39
+ ],
40
+ "image_processor_type": "LlavaNextImageProcessor",
41
+ "image_std": [
42
+ 0.26862954,
43
+ 0.26130258,
44
+ 0.27577711
45
+ ],
46
+ "processor_class": "LlavaNextProcessor",
47
+ "resample": 3,
48
+ "rescale_factor": 0.00392156862745098,
49
+ "size": {
50
+ "shortest_edge": 336
51
+ }
52
+ }
slice_36624/processor_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_token": "<image>",
3
+ "num_additional_image_tokens": 1,
4
+ "patch_size": 14,
5
+ "processor_class": "LlavaNextProcessor",
6
+ "vision_feature_select_strategy": "default"
7
+ }
slice_36624/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d71fca0b80d0ad8f53fa03d6f5208cf203f94fbb7d9672908b0e8e02dcb4323
3
+ size 15133733934
slice_36624/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "image_token": "<image>",
17
+ "pad_token": {
18
+ "content": "<pad>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "unk_token": {
25
+ "content": "<unk>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
slice_36624/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
slice_36624/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
slice_36624/tokenizer_config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "32000": {
31
+ "content": "<image>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "32001": {
39
+ "content": "<pad>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ }
46
+ },
47
+ "additional_special_tokens": [],
48
+ "bos_token": "<s>",
49
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
50
+ "clean_up_tokenization_spaces": false,
51
+ "eos_token": "</s>",
52
+ "extra_special_tokens": {
53
+ "image_token": "<image>"
54
+ },
55
+ "image_token": "<image>",
56
+ "legacy": true,
57
+ "max_length": null,
58
+ "model_max_length": 4096,
59
+ "pad_to_multiple_of": null,
60
+ "pad_token": "<pad>",
61
+ "pad_token_type_id": 0,
62
+ "padding_side": "right",
63
+ "processor_class": "LlavaNextProcessor",
64
+ "sp_model_kwargs": {},
65
+ "spaces_between_special_tokens": false,
66
+ "tokenizer_class": "LlamaTokenizer",
67
+ "unk_token": "<unk>",
68
+ "use_default_system_prompt": false
69
+ }
slice_48832/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<image>": 32000,
3
+ "<pad>": 32001
4
+ }
slice_48832/chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% for message in messages %}{% if message['role'] != 'system' %}{{ message['role'].upper() + ': '}}{% endif %}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>\n' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ content['text'] + ' '}}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ content['text'] + ' '}}{% endgeneration %}{% endfor %}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'ASSISTANT:' }}{% endif %}"
3
+ }
slice_48832/config.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "LlavaNextForConditionalGeneration"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "ignore_index": -100,
9
+ "image_grid_pinpoints": [
10
+ [
11
+ 336,
12
+ 672
13
+ ],
14
+ [
15
+ 672,
16
+ 336
17
+ ],
18
+ [
19
+ 672,
20
+ 672
21
+ ],
22
+ [
23
+ 1008,
24
+ 336
25
+ ],
26
+ [
27
+ 336,
28
+ 1008
29
+ ]
30
+ ],
31
+ "image_seq_length": 576,
32
+ "image_token_index": 32000,
33
+ "model_type": "llava_next",
34
+ "multimodal_projector_bias": true,
35
+ "pad_token_id": 32001,
36
+ "projector_hidden_act": "gelu",
37
+ "text_config": {
38
+ "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
39
+ "architectures": [
40
+ "MistralForCausalLM"
41
+ ],
42
+ "attention_dropout": 0.0,
43
+ "head_dim": 128,
44
+ "hidden_act": "silu",
45
+ "hidden_size": 4096,
46
+ "initializer_range": 0.02,
47
+ "intermediate_size": 14336,
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "mistral",
50
+ "num_attention_heads": 32,
51
+ "num_hidden_layers": 32,
52
+ "num_key_value_heads": 8,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_theta": 1000000.0,
55
+ "sliding_window": null,
56
+ "torch_dtype": "bfloat16",
57
+ "use_cache": true,
58
+ "vocab_size": 32064
59
+ },
60
+ "tie_word_embeddings": false,
61
+ "torch_dtype": "bfloat16",
62
+ "transformers_version": "4.50.0",
63
+ "use_image_newline_parameter": true,
64
+ "vision_config": {
65
+ "attention_dropout": 0.0,
66
+ "hidden_act": "quick_gelu",
67
+ "hidden_size": 1024,
68
+ "image_size": 336,
69
+ "initializer_factor": 1.0,
70
+ "initializer_range": 0.02,
71
+ "intermediate_size": 4096,
72
+ "layer_norm_eps": 1e-05,
73
+ "model_type": "clip_vision_model",
74
+ "num_attention_heads": 16,
75
+ "num_channels": 3,
76
+ "num_hidden_layers": 24,
77
+ "patch_size": 14,
78
+ "projection_dim": 768,
79
+ "torch_dtype": "bfloat16",
80
+ "vocab_size": 32000
81
+ },
82
+ "vision_feature_layer": -2,
83
+ "vision_feature_select_strategy": "default",
84
+ "vocab_size": 32064
85
+ }
slice_48832/preprocessor_config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "aspect_ratio_setting": "anyres",
3
+ "crop_size": {
4
+ "height": 336,
5
+ "width": 336
6
+ },
7
+ "do_center_crop": true,
8
+ "do_convert_rgb": true,
9
+ "do_normalize": true,
10
+ "do_pad": true,
11
+ "do_rescale": true,
12
+ "do_resize": true,
13
+ "image_grid_pinpoints": [
14
+ [
15
+ 336,
16
+ 672
17
+ ],
18
+ [
19
+ 672,
20
+ 336
21
+ ],
22
+ [
23
+ 672,
24
+ 672
25
+ ],
26
+ [
27
+ 1008,
28
+ 336
29
+ ],
30
+ [
31
+ 336,
32
+ 1008
33
+ ]
34
+ ],
35
+ "image_mean": [
36
+ 0.48145466,
37
+ 0.4578275,
38
+ 0.40821073
39
+ ],
40
+ "image_processor_type": "LlavaNextImageProcessor",
41
+ "image_std": [
42
+ 0.26862954,
43
+ 0.26130258,
44
+ 0.27577711
45
+ ],
46
+ "processor_class": "LlavaNextProcessor",
47
+ "resample": 3,
48
+ "rescale_factor": 0.00392156862745098,
49
+ "size": {
50
+ "shortest_edge": 336
51
+ }
52
+ }
slice_48832/processor_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_token": "<image>",
3
+ "num_additional_image_tokens": 1,
4
+ "patch_size": 14,
5
+ "processor_class": "LlavaNextProcessor",
6
+ "vision_feature_select_strategy": "default"
7
+ }
slice_48832/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c7bc2f53054ae607fc6ed062ed63edc9549dcd56a12277d6d5a2f60cf98e8c8
3
+ size 15133733934
slice_48832/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "image_token": "<image>",
17
+ "pad_token": {
18
+ "content": "<pad>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "unk_token": {
25
+ "content": "<unk>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
slice_48832/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
slice_48832/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
slice_48832/tokenizer_config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "32000": {
31
+ "content": "<image>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "32001": {
39
+ "content": "<pad>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ }
46
+ },
47
+ "additional_special_tokens": [],
48
+ "bos_token": "<s>",
49
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
50
+ "clean_up_tokenization_spaces": false,
51
+ "eos_token": "</s>",
52
+ "extra_special_tokens": {
53
+ "image_token": "<image>"
54
+ },
55
+ "image_token": "<image>",
56
+ "legacy": true,
57
+ "max_length": null,
58
+ "model_max_length": 4096,
59
+ "pad_to_multiple_of": null,
60
+ "pad_token": "<pad>",
61
+ "pad_token_type_id": 0,
62
+ "padding_side": "right",
63
+ "processor_class": "LlavaNextProcessor",
64
+ "sp_model_kwargs": {},
65
+ "spaces_between_special_tokens": false,
66
+ "tokenizer_class": "LlamaTokenizer",
67
+ "unk_token": "<unk>",
68
+ "use_default_system_prompt": false
69
+ }
slice_61040/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<image>": 32000,
3
+ "<pad>": 32001
4
+ }
slice_61040/chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% for message in messages %}{% if message['role'] != 'system' %}{{ message['role'].upper() + ': '}}{% endif %}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>\n' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ content['text'] + ' '}}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ content['text'] + ' '}}{% endgeneration %}{% endfor %}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'ASSISTANT:' }}{% endif %}"
3
+ }
slice_61040/config.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "LlavaNextForConditionalGeneration"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "ignore_index": -100,
9
+ "image_grid_pinpoints": [
10
+ [
11
+ 336,
12
+ 672
13
+ ],
14
+ [
15
+ 672,
16
+ 336
17
+ ],
18
+ [
19
+ 672,
20
+ 672
21
+ ],
22
+ [
23
+ 1008,
24
+ 336
25
+ ],
26
+ [
27
+ 336,
28
+ 1008
29
+ ]
30
+ ],
31
+ "image_seq_length": 576,
32
+ "image_token_index": 32000,
33
+ "model_type": "llava_next",
34
+ "multimodal_projector_bias": true,
35
+ "pad_token_id": 32001,
36
+ "projector_hidden_act": "gelu",
37
+ "text_config": {
38
+ "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
39
+ "architectures": [
40
+ "MistralForCausalLM"
41
+ ],
42
+ "attention_dropout": 0.0,
43
+ "head_dim": 128,
44
+ "hidden_act": "silu",
45
+ "hidden_size": 4096,
46
+ "initializer_range": 0.02,
47
+ "intermediate_size": 14336,
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "mistral",
50
+ "num_attention_heads": 32,
51
+ "num_hidden_layers": 32,
52
+ "num_key_value_heads": 8,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_theta": 1000000.0,
55
+ "sliding_window": null,
56
+ "torch_dtype": "bfloat16",
57
+ "use_cache": true,
58
+ "vocab_size": 32064
59
+ },
60
+ "tie_word_embeddings": false,
61
+ "torch_dtype": "bfloat16",
62
+ "transformers_version": "4.50.0",
63
+ "use_image_newline_parameter": true,
64
+ "vision_config": {
65
+ "attention_dropout": 0.0,
66
+ "hidden_act": "quick_gelu",
67
+ "hidden_size": 1024,
68
+ "image_size": 336,
69
+ "initializer_factor": 1.0,
70
+ "initializer_range": 0.02,
71
+ "intermediate_size": 4096,
72
+ "layer_norm_eps": 1e-05,
73
+ "model_type": "clip_vision_model",
74
+ "num_attention_heads": 16,
75
+ "num_channels": 3,
76
+ "num_hidden_layers": 24,
77
+ "patch_size": 14,
78
+ "projection_dim": 768,
79
+ "torch_dtype": "bfloat16",
80
+ "vocab_size": 32000
81
+ },
82
+ "vision_feature_layer": -2,
83
+ "vision_feature_select_strategy": "default",
84
+ "vocab_size": 32064
85
+ }
slice_61040/preprocessor_config.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "aspect_ratio_setting": "anyres",
3
+ "crop_size": {
4
+ "height": 336,
5
+ "width": 336
6
+ },
7
+ "do_center_crop": true,
8
+ "do_convert_rgb": true,
9
+ "do_normalize": true,
10
+ "do_pad": true,
11
+ "do_rescale": true,
12
+ "do_resize": true,
13
+ "image_grid_pinpoints": [
14
+ [
15
+ 336,
16
+ 672
17
+ ],
18
+ [
19
+ 672,
20
+ 336
21
+ ],
22
+ [
23
+ 672,
24
+ 672
25
+ ],
26
+ [
27
+ 1008,
28
+ 336
29
+ ],
30
+ [
31
+ 336,
32
+ 1008
33
+ ]
34
+ ],
35
+ "image_mean": [
36
+ 0.48145466,
37
+ 0.4578275,
38
+ 0.40821073
39
+ ],
40
+ "image_processor_type": "LlavaNextImageProcessor",
41
+ "image_std": [
42
+ 0.26862954,
43
+ 0.26130258,
44
+ 0.27577711
45
+ ],
46
+ "processor_class": "LlavaNextProcessor",
47
+ "resample": 3,
48
+ "rescale_factor": 0.00392156862745098,
49
+ "size": {
50
+ "shortest_edge": 336
51
+ }
52
+ }
slice_61040/processor_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_token": "<image>",
3
+ "num_additional_image_tokens": 1,
4
+ "patch_size": 14,
5
+ "processor_class": "LlavaNextProcessor",
6
+ "vision_feature_select_strategy": "default"
7
+ }
slice_61040/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9595744c46c595c10945c3a98cdbbaec4d93aee2426da2e70482287143c244fb
3
+ size 15133733934