对部分示例程序添加对unsloth的支持

.gitignore

@@ -25,7 +25,7 @@ share/python-wheels/
 .installed.cfg
 *.egg
 MANIFEST
-
+unsloth_compiled_cache/
 # PyInstaller
 #  Usually these files are written by a python script from a template
 #  before PyInstaller builds the stand-alone binary.

=0.9.3

@@ -0,0 +1,5 @@
+Collecting deepspeed
+  Downloading deepspeed-0.19.2.tar.gz (1.8 MB)
+     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.8/1.8 MB 824.3 kB/s  0:00:02
+  Installing build dependencies: started
+  Installing build dependencies: finished with status 'canceled'

bashrc1

@@ -0,0 +1,133 @@
+# ~/.bashrc: executed by bash(1) for non-login shells.
+# see /usr/share/doc/bash/examples/startup-files (in the package bash-doc)
+# for examples
+
+# If not running interactively, don't do anything
+[ -z "$PS1" ] && return
+
+# don't put duplicate lines in the history. See bash(1) for more options
+# ... or force ignoredups and ignorespace
+HISTCONTROL=ignoredups:ignorespace
+
+# append to the history file, don't overwrite it
+shopt -s histappend
+
+# for setting history length see HISTSIZE and HISTFILESIZE in bash(1)
+HISTSIZE=1000
+HISTFILESIZE=2000
+
+# check the window size after each command and, if necessary,
+# update the values of LINES and COLUMNS.
+shopt -s checkwinsize
+
+# make less more friendly for non-text input files, see lesspipe(1)
+[ -x /usr/bin/lesspipe ] && eval "$(SHELL=/bin/sh lesspipe)"
+
+# set variable identifying the chroot you work in (used in the prompt below)
+if [ -z "$debian_chroot" ] && [ -r /etc/debian_chroot ]; then
+    debian_chroot=$(cat /etc/debian_chroot)
+fi
+
+# set a fancy prompt (non-color, unless we know we "want" color)
+case "$TERM" in
+    xterm-color) color_prompt=yes;;
+esac
+
+# uncomment for a colored prompt, if the terminal has the capability; turned
+# off by default to not distract the user: the focus in a terminal window
+# should be on the output of commands, not on the prompt
+#force_color_prompt=yes
+
+if [ -n "$force_color_prompt" ]; then
+    if [ -x /usr/bin/tput ] && tput setaf 1 >&/dev/null; then
+	# We have color support; assume it's compliant with Ecma-48
+	# (ISO/IEC-6429). (Lack of such support is extremely rare, and such
+	# a case would tend to support setf rather than setaf.)
+	color_prompt=yes
+    else
+	color_prompt=
+    fi
+fi
+
+if [ "$color_prompt" = yes ]; then
+    PS1='${debian_chroot:+($debian_chroot)}\[\033[01;32m\]\u@\h\[\033[00m\]:\[\033[01;34m\]\w\[\033[00m\]\$ '
+else
+    PS1='${debian_chroot:+($debian_chroot)}\u@\h:\w\$ '
+fi
+unset color_prompt force_color_prompt
+
+# If this is an xterm set the title to user@host:dir
+case "$TERM" in
+xterm*|rxvt*)
+    PS1="\[\e]0;${debian_chroot:+($debian_chroot)}\u@\h: \w\a\]$PS1"
+    ;;
+*)
+    ;;
+esac
+
+# enable color support of ls and also add handy aliases
+if [ -x /usr/bin/dircolors ]; then
+    test -r ~/.dircolors && eval "$(dircolors -b ~/.dircolors)" || eval "$(dircolors -b)"
+    alias ls='ls --color=auto'
+    #alias dir='dir --color=auto'
+    #alias vdir='vdir --color=auto'
+
+    alias grep='grep --color=auto'
+    alias fgrep='fgrep --color=auto'
+    alias egrep='egrep --color=auto'
+fi
+
+# some more ls aliases
+alias ll='ls -alF'
+alias la='ls -A'
+alias l='ls -CF'
+
+# Alias definitions.
+# You may want to put all your additions into a separate file like
+# ~/.bash_aliases, instead of adding them here directly.
+# See /usr/share/doc/bash-doc/examples in the bash-doc package.
+
+if [ -f ~/.bash_aliases ]; then
+    . ~/.bash_aliases
+fi
+
+# enable programmable completion features (you don't need to enable
+# this, if it's already enabled in /etc/bash.bashrc and /etc/profile
+# sources /etc/bash.bashrc).
+#if [ -f /etc/bash_completion ] && ! shopt -oq posix; then
+#    . /etc/bash_completion
+#fi
+export PS1="[\u:\W]\$ "
+
+# INSPIRE_PATH_CONFIG
+export PATH=$PATH:"/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/inspire/hdd/global_public/public_bin"
+if [ -f ${HOME}/.welcome_Inspire ]; then eval "echo -e \"$(cat ${HOME}/.welcome_Inspire)\""; fi
+
+# >>> conda initialize >>>
+# !! Contents within this block are managed by 'conda init' !!
+__conda_setup="$('/inspire/hdd/global_user/shixiaoxin-253107030017/sxx/cialtion_train/miniconda/bin/conda' 'shell.bash' 'hook' 2> /dev/null)"
+if [ $? -eq 0 ]; then
+    eval "$__conda_setup"
+else
+    if [ -f "/inspire/hdd/global_user/shixiaoxin-253107030017/sxx/cialtion_train/miniconda/etc/profile.d/conda.sh" ]; then
+        . "/inspire/hdd/global_user/shixiaoxin-253107030017/sxx/cialtion_train/miniconda/etc/profile.d/conda.sh"
+    else
+        export PATH="/inspire/hdd/global_user/shixiaoxin-253107030017/sxx/cialtion_train/miniconda/bin:$PATH"
+    fi
+fi
+unset __conda_setup
+# <<< conda initialize <<<
+
+
+# >>> mamba initialize >>>
+# !! Contents within this block are managed by 'mamba shell init' !!
+export MAMBA_EXE='/inspire/hdd/global_user/shixiaoxin-253107030017/sxx/cialtion_train/miniconda/bin/mamba';
+export MAMBA_ROOT_PREFIX='/root/.local/share/mamba';
+__mamba_setup="$("$MAMBA_EXE" shell hook --shell bash --root-prefix "$MAMBA_ROOT_PREFIX" 2> /dev/null)"
+if [ $? -eq 0 ]; then
+    eval "$__mamba_setup"
+else
+    alias mamba="$MAMBA_EXE"  # Fallback on help from mamba activate
+fi
+unset __mamba_setup
+# <<< mamba initialize <<<

examples/gemma4_engram_lora.py

@@ -400,7 +400,17 @@ def get_vocab_size(self: Any) -> int | None:
     # KV cache during generation and causes repeated/garbled output.
     model.gradient_checkpointing_disable()
     print("\n>>> Inference Demo (Original Model)")
-    prompt = "<start_of_turn>user\nTell me a short fact about the moon.<end_of_turn>\n<start_of_turn>model\n"
+    messages = [{"role": "user", "content": "Tell me a short fact about the moon."}]
+    prompt = tokenizer.apply_chat_template(
+        messages, 
+        tokenize=False, 
+        add_generation_prompt=True,
+        # enable_thinking=False
+    )
+
+    #optional↓
+    # prompt = "<|turn>user\nTell me a short fact about the moon.<turn|>\n<|turn>model\n"
+
     # Use hasattr to get the device safely
     if hasattr(model.base_model, "device"):
         target_device = model.base_model.device
@@ -417,7 +427,7 @@ def get_vocab_size(self: Any) -> int | None:
             max_new_tokens=100,
             max_length=None,
             do_sample=False,
-            stop_strings=["<end_of_turn>"],
+            stop_strings=["<turn|>"],
             tokenizer=tokenizer,
         )
     print(f"Response: {tokenizer.decode(output[0], skip_special_tokens=True)}")
@@ -488,4 +498,4 @@ def get_vocab_size(self: Any) -> int | None:
     )
     args = parser.parse_args()
 
-    run_example(args)
+    run_example(args)