build-small-hackathon/Nemotron-nano-4b-escape-room
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: build-small-hackathon/Nemotron-nano-4b-escape-room
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-07-23 22:24:15 +08:00
commit d4b59a67e3
16 changed files with 10586 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
.gitattributes vendored Normal file
View File

@@ -0,0 +1,44 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bz2 filter=lfs diff=lfs merge=lfs -text
*.ckpt filter=lfs diff=lfs merge=lfs -text
*.ftz filter=lfs diff=lfs merge=lfs -text
*.gz filter=lfs diff=lfs merge=lfs -text
*.h5 filter=lfs diff=lfs merge=lfs -text
*.joblib filter=lfs diff=lfs merge=lfs -text
*.lfs.* filter=lfs diff=lfs merge=lfs -text
*.mlmodel filter=lfs diff=lfs merge=lfs -text
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack filter=lfs diff=lfs merge=lfs -text
*.npy filter=lfs diff=lfs merge=lfs -text
*.npz filter=lfs diff=lfs merge=lfs -text
*.onnx filter=lfs diff=lfs merge=lfs -text
*.ot filter=lfs diff=lfs merge=lfs -text
*.parquet filter=lfs diff=lfs merge=lfs -text
*.pb filter=lfs diff=lfs merge=lfs -text
*.pickle filter=lfs diff=lfs merge=lfs -text
*.pkl filter=lfs diff=lfs merge=lfs -text
*.pt filter=lfs diff=lfs merge=lfs -text
*.pth filter=lfs diff=lfs merge=lfs -text
*.rar filter=lfs diff=lfs merge=lfs -text
*.safetensors filter=lfs diff=lfs merge=lfs -text
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
*.tar.* filter=lfs diff=lfs merge=lfs -text
*.tar filter=lfs diff=lfs merge=lfs -text
*.tflite filter=lfs diff=lfs merge=lfs -text
*.tgz filter=lfs diff=lfs merge=lfs -text
*.wasm filter=lfs diff=lfs merge=lfs -text
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip filter=lfs diff=lfs merge=lfs -text
*.zst filter=lfs diff=lfs merge=lfs -text
*tfevents* filter=lfs diff=lfs merge=lfs -text
nemotron-room-lora-fp16.gguf filter=lfs diff=lfs merge=lfs -text
nemotron-room-lora-q4-k-m.gguf filter=lfs diff=lfs merge=lfs -text
nemotron-room-lora-q8-0.gguf filter=lfs diff=lfs merge=lfs -text
nemotron-room-lora-Q4-K-M.gguf filter=lfs diff=lfs merge=lfs -text
nemotron-room-lora-FP16.gguf filter=lfs diff=lfs merge=lfs -text
nemotron-room-lora-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
nemotron-room-lora-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
nemotron-room-lora-F16.gguf filter=lfs diff=lfs merge=lfs -text
tokenizer.json filter=lfs diff=lfs merge=lfs -text

75
README.md Normal file
View File

@@ -0,0 +1,75 @@
---
license: other
license_name: nvidia-nemotron-open-model-license
license_link: >-
https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-nemotron-open-model-license/
tags:
- text-generation
- finetuned
- nemotron
- llama-cpp
- GGUF
language:
- en
base_model: unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF
datasets:
- build-small-hackathon/1000-Rooms-DS
---
# Fine-tuned Nemotron-3 Nano 4B
This model is a fine-tuned version of **NVIDIA Nemotron-3 Nano 4B**, adapted for improved performance on a custom dataset for generating escape rooms. This fine-tune was realised for the build-small-hackathon using **Unsloth** and using compute credits provided by **Modal**.
---
## Model Details
- **Base model:** https://huggingface.co/nvidia/Nemotron-3-Nano-4B
- **Quantized model:** https://huggingface.co/unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF
- **Architecture:** Decoder-only Transformer
- **Parameters:** ~4B
- **Fine-tuning method:** LoRA
- **Framework:** Unsloth
- **Training environment:** Modal
---
## Intended Use
This model is intended for:
- Experimental / research use
- Generating room titles and descriptions / generating responses for trying to open doors/containers as json
It has not been optimized for production safety or alignment.
---
## Limitations
- May produce incorrect or hallucinated outputs
- Not guaranteed to follow instructions perfectly
- Sensitive to prompt formatting
- May inherit biases from base model and training data
---
## How to Use
### llama.cpp
```bat
llama-cli.exe -m nemotron-room-lora-q4-k-m.gguf ^
-n 2048 --temp 0.7 --repeat-penalty 1.1 ^
--grammar-file json.gbnf ^
-sys "You are a creative dungeon-master AI that generates escape-room content. Always respond with a single valid JSON object and nothing else." ^
-p "{\"task\":\"generate_room\"} Respond with room_name, room_story, room_prompt, door_description, door_prompt, door_key_name, door_key_prompt, containers array, and keys array."
```
### Output
```json
{
"room_name": "The Crystal Lake of the Moon", "room_story": "A lake that reflects the moon so perfectly it seems like a second sky. The water is still and silver.", "room_prompt": "A serene lake under a full moon, reflecting the night sky perfectly, surrounded by misty mountains.", "door_description": "The entrance is carved into an ancient oak tree, with vines curling around the frame.", "door_prompt": "An old oak tree door with green vines and moss growing on the wood grain.", "door_key_name": "The Silver Moon Key", "door_key_prompt": "A small silver key shaped like a crescent moon, glowing softly.", "containers": [{"container_name": "The Hollow Oak Chest", "container_prompt": "An old wooden chest carved with animal faces, sitting on mossy ground."}, {"container_name": "The Stone Lantern", "container_prompt": "A large stone lantern containing a bundle of dried herbs and a small scroll."}, {"container_name": "The Leather Satchel", "container_prompt": "A worn leather satchel filled with various trinkets, including a compass and a coin."}, {"container_name": "The Glass Globe", "container_prompt": "A clear glass globe containing swirling blue and silver dust that catches the moonlight."}], "keys": [{"key_name": "The Rusty Key", "key_prompt": "A tarnished iron key with jagged teeth, rusted to a dull orange."}, {"key_name": "The Gold Coin", "key_prompt": "An ornate gold coin embedded with gemstones, resting on a piece of parchment."}]}
```

204
chat_template.jinja Normal file
View File

@@ -0,0 +1,204 @@
{% macro render_extra_keys(json_dict, handled_keys) %}
{%- if json_dict is mapping %}
{%- for json_key in json_dict if json_key not in handled_keys %}
{%- if json_dict[json_key] is mapping or (json_dict[json_key] is sequence and json_dict[json_key] is not string) %}
{{- '\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | tojson ) ~ '</' ~ json_key ~ '>' }}
{%- else %}
{{-'\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | string) ~ '</' ~ json_key ~ '>' }}
{%- endif %}
{%- endfor %}
{%- endif %}
{% endmacro %}
{%- set enable_thinking = enable_thinking if enable_thinking is defined else True %}
{%- set truncate_history_thinking = truncate_history_thinking if truncate_history_thinking is defined else True %}
{%- set ns = namespace(last_user_idx = -1) %}
{%- set loop_messages = messages %}
{%- for m in loop_messages %}
{%- if m["role"] == "user" %}
{%- set ns.last_user_idx = loop.index0 %}
{%- endif %}
{%- endfor %}
{%- if messages[0]["role"] == "system" %}
{%- set system_message = messages[0]["content"] %}
{%- set loop_messages = messages[1:] %}
{%- else %}
{%- set system_message = "" %}
{%- set loop_messages = messages %}
{%- endif %}
{%- if not tools is defined %}
{%- set tools = [] %}
{%- endif %}
{# Recompute last_user_idx relative to loop_messages after handling system #}
{%- set ns = namespace(last_user_idx = -1) %}
{%- for m in loop_messages %}
{%- if m["role"] == "user" %}
{%- set ns.last_user_idx = loop.index0 %}
{%- endif %}
{%- endfor %}
{%- if system_message is defined %}
{{- "<|im_start|>system\n" + system_message }}
{%- else %}
{%- if tools is iterable and tools | length > 0 %}
{{- "<|im_start|>system\n" }}
{%- endif %}
{%- endif %}
{%- if tools is iterable and tools | length > 0 %}
{%- if system_message is defined and system_message | length > 0 %}
{{- "\n\n" }}
{%- endif %}
{{- "# Tools\n\nYou have access to the following functions:\n\n" }}
{{- "<tools>" }}
{%- for tool in tools %}
{%- if tool.function is defined %}
{%- set tool = tool.function %}
{%- endif %}
{{- "\n<function>\n<name>" ~ tool.name ~ "</name>" }}
{%- if tool.description is defined %}
{{- '\n<description>' ~ (tool.description | trim) ~ '</description>' }}
{%- endif %}
{{- '\n<parameters>' }}
{%- if tool.parameters is defined and tool.parameters is mapping and tool.parameters.properties is defined and tool.parameters.properties is mapping %}
{%- for param_name, param_fields in tool.parameters.properties|items %}
{{- '\n<parameter>' }}
{{- '\n<name>' ~ param_name ~ '</name>' }}
{%- if param_fields.type is defined %}
{{- '\n<type>' ~ (param_fields.type | string) ~ '</type>' }}
{%- endif %}
{%- if param_fields.description is defined %}
{{- '\n<description>' ~ (param_fields.description | trim) ~ '</description>' }}
{%- endif %}
{%- if param_fields.enum is defined %}
{{- '\n<enum>' ~ (param_fields.enum | tojson ) ~ '</enum>' }}
{%- endif %}
{%- set handled_keys = ['name', 'type', 'description', 'enum'] %}
{{- render_extra_keys(param_fields, handled_keys) }}
{{- '\n</parameter>' }}
{%- endfor %}
{%- endif %}
{% set handled_keys = ['type', 'properties', 'required'] %}
{{- render_extra_keys(tool.parameters, handled_keys) }}
{%- if tool.parameters is defined and tool.parameters.required is defined %}
{{- '\n<required>' ~ (tool.parameters.required | tojson ) ~ '</required>' }}
{%- endif %}
{{- '\n</parameters>' }}
{%- set handled_keys = ['type', 'name', 'description', 'parameters'] %}
{{- render_extra_keys(tool, handled_keys) }}
{{- '\n</function>' }}
{%- endfor %}
{{- "\n</tools>" }}
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
{%- endif %}
{%- if system_message is defined %}
{{- '<|im_end|>\n' }}
{%- else %}
{%- if tools is iterable and tools | length > 0 %}
{{- '<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- for message in loop_messages %}
{%- if message.role == "assistant" %}
{# Add reasoning content in to content field for unified processing below. #}
{%- if message.reasoning_content is defined and message.reasoning_content is string and message.reasoning_content | trim | length > 0 %}
{%- set content = "<think>\n" ~ message.reasoning_content ~ "\n</think>\n" ~ (message.content | default('', true)) %}
{%- else %}
{%- set content = message.content | default('', true) %}
{%- if content is string -%}
{# Allow downstream logic to to take care of broken thought, only handle coherent reasoning here. #}
{%- if '<think>' not in content and '</think>' not in content -%}
{%- set content = "<think></think>" ~ content -%}
{%- endif -%}
{%- else -%}
{%- set content = content -%}
{%- endif -%}
{%- endif %}
{%- if message.tool_calls is defined and message.tool_calls is iterable and message.tool_calls | length > 0 %}
{# Assistant message has tool calls. #}
{{- '<|im_start|>assistant\n' }}
{%- set include_content = not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
{%- if content is string and content | trim | length > 0 %}
{%- if include_content %}
{{- (content | trim) ~ '\n' -}}
{%- else %}
{%- set c = (content | string) %}
{%- if '</think>' in c %}
{# Keep only content after the last closing think. Also generation prompt causes this. #}
{%- set c = c.split('</think>')[-1] %}
{%- elif '<think>' in c %}
{# If <think> was opened but never closed, drop the trailing think segment #}
{%- set c = c.split('<think>')[0] %}
{%- endif %}
{%- set c = "<think></think>" ~ c | trim %}
{%- if c | length > 0 %}
{{- c ~ '\n' -}}
{%- endif %}
{%- endif %}
{%- else %}
{{- "<think></think>" -}}
{%- endif %}
{%- for tool_call in message.tool_calls %}
{%- if tool_call.function is defined %}
{%- set tool_call = tool_call.function %}
{%- endif %}
{{- '<tool_call>\n<function=' ~ tool_call.name ~ '>\n' -}}
{%- if tool_call.arguments is defined %}
{%- for args_name, args_value in tool_call.arguments|items %}
{{- '<parameter=' ~ args_name ~ '>\n' -}}
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
{{- args_value ~ '\n</parameter>\n' -}}
{%- endfor %}
{%- endif %}
{{- '</function>\n</tool_call>\n' -}}
{%- endfor %}
{{- '<|im_end|>\n' }}
{%- else %}
{# Assistant message doesn't have tool calls. #}
{%- if not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
{{- '<|im_start|>assistant\n' ~ (content | default('', true) | string | trim) ~ '<|im_end|>\n' }}
{%- else %}
{%- set c = (content | default('', true) | string) %}
{%- if '<think>' in c and '</think>' in c %}
{%- set c = "<think></think>" ~ c.split('</think>')[-1] %}
{%- endif %}
{%- set c = c | trim %}
{%- if c | length > 0 %}
{{- '<|im_start|>assistant\n' ~ c ~ '<|im_end|>\n' }}
{%- else %}
{{- '<|im_start|>assistant\n<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- endif %}
{%- elif message.role == "user" or message.role == "system" %}
{{- '<|im_start|>' + message.role + '\n' }}
{%- set content = message.content | string %}
{{- content }}
{{- '<|im_end|>\n' }}
{%- elif message.role == "tool" %}
{%- if loop.previtem and loop.previtem.role != "tool" %}
{{- '<|im_start|>user\n' }}
{%- endif %}
{{- '<tool_response>\n' }}
{{- message.content }}
{{- '\n</tool_response>\n' }}
{%- if not loop.last and loop.nextitem.role != "tool" %}
{{- '<|im_end|>\n' }}
{%- elif loop.last %}
{{- '<|im_end|>\n' }}
{%- endif %}
{%- else %}
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>\n' }}
{%- endif %}
{%- endfor %}
{%- if add_generation_prompt %}
{%- if enable_thinking %}
{{- '<|im_start|>assistant\n<think>\n' }}
{%- else %}
{{- '<|im_start|>assistant\n<think></think>' }}
{%- endif %}
{%- endif %}

61
config.json Normal file
View File

@@ -0,0 +1,61 @@
{
"architectures": [
"NemotronHForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"attention_head_dim": 128,
"auto_map": {
"AutoConfig": "configuration_nemotron_h.NemotronHConfig",
"AutoModelForCausalLM": "modeling_nemotron_h.NemotronHForCausalLM"
},
"bos_token_id": 1,
"chunk_size": 256,
"conv_kernel": 4,
"torch_dtype": "bfloat16",
"eos_token_id": 11,
"expand": 2,
"head_dim": 128,
"hidden_dropout": 0.0,
"hidden_size": 3136,
"hybrid_override_pattern": "M-M-M-MM-M-M*-M-M*-M-M-M*-M-M-MM*-MMM-M-M-",
"initializer_range": 0.02,
"intermediate_size": 12544,
"layer_norm_epsilon": 1e-05,
"mamba_head_dim": 80,
"mamba_hidden_act": "silu",
"mamba_num_heads": 96,
"mamba_proj_bias": false,
"max_position_embeddings": 262144,
"mlp_bias": false,
"mlp_hidden_act": "relu2",
"model_name": "unsloth/NVIDIA-Nemotron-3-Nano-4B",
"model_type": "nemotron_h",
"n_groups": 8,
"num_attention_heads": 40,
"num_hidden_layers": 42,
"num_key_value_heads": 8,
"num_logits_to_keep": 1,
"pad_token_id": 999,
"rescale_prenorm_residual": true,
"residual_in_fp32": false,
"rms_norm_eps": 1e-05,
"sliding_window": null,
"ssm_state_size": 128,
"tie_word_embeddings": false,
"time_step_floor": 0.0001,
"time_step_limit": [
0.0,
Infinity
],
"time_step_max": 0.1,
"time_step_min": 0.001,
"time_step_rank": 256,
"unsloth_fixed": true,
"unsloth_version": "2026.6.1",
"use_bias": false,
"use_cache": true,
"use_conv_bias": true,
"use_mamba_kernels": true,
"vocab_size": 131072
}

243
configuration_nemotron_h.py Normal file
View File

@@ -0,0 +1,243 @@
# coding=utf-8
# Copyright 2024 AI21 Labs Ltd. and the HuggingFace Inc. team. All rights reserved.
# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""NemotronH model configuration"""
import re
from transformers.configuration_utils import PretrainedConfig
from transformers.utils import logging
logger = logging.get_logger(__name__)
class NemotronHConfig(PretrainedConfig):
r"""
This is the configuration class to store the configuration of a [`NemotronHModel`]. It is used to instantiate a
NemotronH model according to the specified arguments, defining the model architecture. Instantiating a configuration
with the defaults will yield a similar configuration to that of the NemotronH-v0.1 model.
[todo](todo)
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.
Args:
vocab_size (`int`, *optional*, defaults to 131072):
Vocabulary size of the NemotronH model. Defines the number of different tokens that can be represented by the
`inputs_ids` passed when calling [`NemotronHModel`]
tie_word_embeddings (`bool`, *optional*, defaults to `False`):
Whether the model's input and output word embeddings should be tied. Note that this is only relevant if the
model has a output word embedding layer.
hidden_size (`int`, *optional*, defaults to 4096):
Dimension of the hidden representations.
intermediate_size (`int`, *optional*, defaults to 21504):
Dimension of the MLP representations.
num_hidden_layers (`int`, *optional*, defaults to 52):
Number of hidden layers in the Transformer encoder.
hybrid_override_pattern (`str`, *optional*, defaults to `"M-M-M-M*-M-M-M-M-M*-M-M-M-M-M*-M-M-M-M-M*-M-M-M-M-M-"`):
The pattern of the hybrid model. The pattern is a string of characters where each character represents M: Mamba2, *: Attention, -: MLP
num_attention_heads (`int`, *optional*, defaults to 32):
Number of attention heads for each attention layer in the Transformer encoder.
attention_head_dim (`int`, *optional*, defaults to 128):
Dimension of each attention head.
num_key_value_heads (`int`, *optional*, defaults to 8):
This is the number of key_value heads that should be used to implement Grouped Query Attention. If
`num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
`num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used.
mlp_hidden_act (`str`, *optional*, defaults to "relu2"):
The non-linear activation function in the MLP layers.
attention_bias (`bool`, *optional*, defaults to `False`):
Whether to use bias in attention layers.
mlp_bias (`bool`, *optional*, defaults to `False`):
Whether to use bias in MLP layers.
use_bias (`bool`, *optional*, defaults to `False`):
Whether to use bias in the model.
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
layer_norm_epsilon (`float`, *optional*, defaults to 1e-5):
The epsilon used by the layer normalization layers.
residual_in_fp32 (`bool`, *optional*, defaults to `False`):
Whether or not residuals should be in `float32`. If set to `False` residuals will keep the same `dtype` as the rest of the model.
use_cache (`bool`, *optional*, defaults to `True`):
Whether or not the model should return the last key/values attentions (not used by all models). Only
relevant if `config.is_decoder=True`.
num_logits_to_keep (`int` or `None`, *optional*, defaults to 1):
Number of prompt logits to calculate during generation. If `None`, all logits will be calculated. If an
integer value, only last `num_logits_to_keep` logits will be calculated.
pad_token_id (`int`, *optional*, defaults to 0):
The id of the padding token.
bos_token_id (`int`, *optional*, defaults to 1):
The id of the "beginning-of-sequence" token.
eos_token_id (`int`, *optional*, defaults to 2):
The id of the "end-of-sequence" token.
sliding_window (`int`, *optional*, defaults to None):
Sliding window attention window size.
max_position_embeddings (`int`, *optional*, defaults to 4096):
The maximum sequence length that this model might ever be used with.
attention_dropout (`float`, *optional*, defaults to 0.0):
The dropout ratio for the attention probabilities.
hidden_dropout (`float`, *optional*, defaults to 0.0):
The dropout ratio for the hidden states.
use_mamba_kernels (`bool`, *optional*, defaults to `True`):
Flag indicating whether or not to use the fast mamba kernels. These are available only if `mamba-ssm` and
`causal-conv1d` are installed, and the mamba modules are running on a CUDA device.
ssm_state_size (`int`, *optional*, defaults to 128):
The dimension of the mamba state space latents.
mamba_num_heads (`int`, *optional*, defaults to 128):
Number of heads in Mamba layers.
mamba_n_groups (`int`, *optional*, defaults to 8):
Number of groups in Mamba layers.
mamba_head_dim (`int`, *optional*, defaults to 64):
Dimension of each Mamba head.
mamba_d_conv (`int`, *optional*, defaults to 4):
The size of the mamba convolution kernel.
mamba_expand (`int`, *optional*, defaults to 2):
Expanding factor used to determine the mamba intermediate size.
mamba_hidden_act (`str`, *optional*, defaults to "silu"):
The non-linear activation function in the Mamba layers.
mamba_dt_min (`float`, *optional*, defaults to 0.001):
Minimum value for the time step in Mamba.
mamba_dt_max (`float`, *optional*, defaults to 0.1):
Maximum value for the time step in Mamba.
mamba_dt_limit (`tuple`, *optional*, defaults to (0.0, float("inf"))):
Limits for the time step in Mamba.
mamba_dt_init_floor (`float`, *optional*, defaults to 1e-4):
Floor value for time step initialization in Mamba.
mamba_conv_bias (`bool`, *optional*, defaults to `True`):
Whether to use bias in the convolution layer of the mamba mixer block.
mamba_proj_bias (`bool`, *optional*, defaults to `False`):
Whether to use bias in the input and output projections of the mamba mixer block.
mamba_chunk_size (`int`, *optional*, defaults to 256):
Size of chunks for Mamba processing.
rescale_prenorm_residual (`bool`, *optional*, defaults to `True`):
Whether to rescale the pre-normalization residual connections.
"""
model_type = "nemotron_h"
keys_to_ignore_at_inference = ["past_key_values"]
def __init__(
self,
vocab_size=131072,
tie_word_embeddings=False,
hidden_size=4096,
intermediate_size=21504,
num_hidden_layers=52,
hybrid_override_pattern="M-M-M-M*-M-M-M-M-M*-M-M-M-M-M*-M-M-M-M-M*-M-M-M-M-M-",
num_attention_heads=32,
attention_head_dim=128,
num_key_value_heads=8, # nemo: num_query_groups
mlp_hidden_act="relu2",
attention_bias=False,
mlp_bias=False,
use_bias=False,
initializer_range=0.02, # nemo: init_method_std
layer_norm_epsilon=1e-5, # nemo: layernorm_epsilon
residual_in_fp32=False, # Megatron Core default value
use_cache=True,
num_logits_to_keep=1,
pad_token_id=0,
bos_token_id=1,
eos_token_id=2,
sliding_window=None,
max_position_embeddings=4096,
attention_dropout=0.0,
hidden_dropout=0.0, # * ADDED
use_mamba_kernels=True,
ssm_state_size=128, # mamba_state_size
mamba_num_heads=128,
mamba_n_groups=8, # nemo: mamba_ssm_ngroups = num_heads
mamba_head_dim=64,
mamba_d_conv=4,
mamba_expand=2,
mamba_hidden_act="silu",
mamba_dt_min=0.001,
mamba_dt_max=0.1,
mamba_dt_limit=(0.0, float("inf")),
mamba_dt_init_floor=1e-4,
mamba_conv_bias=True,
mamba_proj_bias=False,
mamba_chunk_size=256,
rescale_prenorm_residual=True,
**kwargs,
):
self.vocab_size = vocab_size
self.tie_word_embeddings = tie_word_embeddings
self.hidden_size = hidden_size
self.intermediate_size = intermediate_size
self.num_hidden_layers = num_hidden_layers
self.hybrid_override_pattern = hybrid_override_pattern
self.num_attention_heads = num_attention_heads
self.attention_head_dim = attention_head_dim
self.sliding_window = sliding_window
self.max_position_embeddings = max_position_embeddings
self.attention_dropout = attention_dropout
self.hidden_dropout = hidden_dropout
# Validate hybrid_override_pattern
# M: Mamba2, *: Attention, -: MLP
assert len(self.hybrid_override_pattern) == self.num_hidden_layers, "hybrid_override_pattern must have the same length as num_hidden_layers"
assert re.match(r"^[*-M]+$", self.hybrid_override_pattern), "hybrid_override_pattern must only contain characters 'M', '*', or '-'"
# for backward compatibility
if num_key_value_heads is None:
num_key_value_heads = num_attention_heads
self.num_key_value_heads = num_key_value_heads
self.mlp_hidden_act = mlp_hidden_act
self.attention_bias = attention_bias
self.mlp_bias = mlp_bias
self.use_bias = use_bias
self.initializer_range = initializer_range
self.layer_norm_epsilon = layer_norm_epsilon
self.residual_in_fp32 = residual_in_fp32
self.use_cache = use_cache
self.num_logits_to_keep = num_logits_to_keep
self.use_mamba_kernels = use_mamba_kernels
self.n_groups = mamba_n_groups
self.mamba_head_dim = mamba_head_dim
self.ssm_state_size = ssm_state_size
self.mamba_num_heads = mamba_num_heads
self.conv_kernel = mamba_d_conv
self.expand = mamba_expand
self.mamba_hidden_act = mamba_hidden_act
self.time_step_min = mamba_dt_min
self.time_step_max = mamba_dt_max
self.time_step_limit = mamba_dt_limit
self.time_step_floor = mamba_dt_init_floor
self.use_conv_bias = mamba_conv_bias
self.mamba_proj_bias = mamba_proj_bias
self.chunk_size = mamba_chunk_size
self.rescale_prenorm_residual = rescale_prenorm_residual
super().__init__(
pad_token_id=pad_token_id,
bos_token_id=bos_token_id,
eos_token_id=eos_token_id,
tie_word_embeddings=tie_word_embeddings,
**kwargs,
)
@property
def layers_block_type(self):
return [
"mamba" if self.hybrid_override_pattern[i] == "M" else
"attention" if self.hybrid_override_pattern[i] == "*" else "mlp"
for i in range(self.num_hidden_layers)]

10
generation_config.json Normal file
View File

@@ -0,0 +1,10 @@
{
"_from_model_config": true,
"bos_token_id": 1,
"eos_token_id": [
11,
2
],
"pad_token_id": 999,
"transformers_version": "4.57.6"
}

24
json.gbnf Normal file
View File

@@ -0,0 +1,24 @@
root ::= object
value ::= object | array | string | number | ("true" | "false" | "null") ws
object ::=
"{" ws (
string ":" ws value
("," ws string ":" ws value)*
)? "}" ws
array ::=
"[" ws (
value
("," ws value)*
)? "]" ws
string ::=
"\"" (
[^\\"\x7F\x00-\x1F] |
"\\" (["\\/bfnrt] | "u" [0-9a-fA-F] [0-9a-fA-F] [0-9a-fA-F] [0-9a-fA-F])
)* "\"" ws
number ::= ("-"? ([0-9] | [1-9] [0-9]*)) ("." [0-9]+)? (([eE] [-+]? [0-9]+))? ws
ws ::= ([ \t\n] ws)?

3
model.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:34ed9c51e2665f0e9a56c68c23f0a17802044d6a43d7aa6dbed6dbbdcd362830
size 7947142640

1638
modeling_nemotron_h.py Normal file
View File

File diff suppressed because it is too large Load Diff

3
nemotron-room-lora-F16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:969b5d3e005ecfee43dbd1af37a0b79695a6d41ec4f1e49c98d4c27ed75af407
size 7957646528

3
nemotron-room-lora-Q4_K_M.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d6f2b5b2c5a23cad474109fa07e8d4d739b95f4bf0ea9cf6acd045f78272f924
size 2837072832

3
nemotron-room-lora-Q8_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:aeb9d9eb72ab9e9397b6c44bd1c07ed940b8e2c7ebea9db11cbc9c21bdd51771
size 4233677888

221
nemotron-template.jinja Normal file
View File

@@ -0,0 +1,221 @@
{# Unsloth template fixes #}
{% macro render_extra_keys(json_dict, handled_keys) %}
{%- if json_dict is mapping %}
{%- for json_key in json_dict if json_key not in handled_keys %}
{%- if json_dict[json_key] is mapping or (json_dict[json_key] is iterable and json_dict[json_key] is not string) %}
{{- '\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | tojson) ~ '</' ~ json_key ~ '>' }}
{%- else %}
{# FIX: Added default('', true) to prevent null errors #}
{{-'\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | default('', true) | string) ~ '</' ~ json_key ~ '>' }}
{%- endif %}
{%- endfor %}
{%- endif %}
{% endmacro %}
{# --- Default is True (Standard behavior) --- #}
{%- set enable_thinking = enable_thinking if enable_thinking is defined else True %}
{%- set truncate_history_thinking = truncate_history_thinking if truncate_history_thinking is defined else True %}
{%- set ns = namespace(last_user_idx = -1) %}
{%- set loop_messages = messages %}
{%- for m in loop_messages %}
{%- if m["role"] == "user" %}
{%- set ns.last_user_idx = loop.index0 %}
{%- endif %}
{%- endfor %}
{%- if messages[0]["role"] == "system" %}
{# FIX: Handle null system content #}
{%- set system_message = messages[0]["content"] | default('', true) %}
{# --- Logic: Check for /nothink trigger --- #}
{%- if '/nothink' in system_message %}
{%- set enable_thinking = False %}
{%- endif %}
{# --------------------------------------------- #}
{%- set loop_messages = messages[1:] %}
{%- else %}
{%- set system_message = "" %}
{%- set loop_messages = messages %}
{%- endif %}
{%- if not tools is defined %}
{%- set tools = [] %}
{%- endif %}
{# Recompute last_user_idx relative to loop_messages after handling system #}
{%- set ns = namespace(last_user_idx = -1) %}
{%- for m in loop_messages %}
{%- if m["role"] == "user" %}
{%- set ns.last_user_idx = loop.index0 %}
{%- endif %}
{%- endfor %}
{%- if system_message is defined %}
{{- "<|im_start|>system\n" + system_message }}
{%- else %}
{%- if tools is iterable and tools | length > 0 %}
{{- "<|im_start|>system\n" }}
{%- endif %}
{%- endif %}
{%- if tools is iterable and tools | length > 0 %}
{%- if system_message is defined and system_message | length > 0 %}
{{- "\n\n" }}
{%- endif %}
{{- "# Tools\n\nYou have access to the following functions:\n\n" }}
{{- "<tools>" }}
{%- for tool in tools %}
{%- if tool.function is defined %}
{%- set tool = tool.function %}
{%- endif %}
{{- "\n<function>\n<name>" ~ tool.name ~ "</name>" }}
{%- if tool.description is defined %}
{{- '\n<description>' ~ (tool.description | trim) ~ '</description>' }}
{%- endif %}
{{- '\n<parameters>' }}
{%- if tool.parameters is defined and tool.parameters is mapping and tool.parameters.properties is defined and tool.parameters.properties is mapping %}
{%- for param_name, param_fields in tool.parameters.properties|items %}
{{- '\n<parameter>' }}
{{- '\n<name>' ~ param_name ~ '</name>' }}
{%- if param_fields.type is defined %}
{{- '\n<type>' ~ (param_fields.type | string) ~ '</type>' }}
{%- endif %}
{%- if param_fields.description is defined %}
{{- '\n<description>' ~ (param_fields.description | trim) ~ '</description>' }}
{%- endif %}
{%- if param_fields.enum is defined %}
{{- '\n<enum>' ~ (param_fields.enum | tojson) ~ '</enum>' }}
{%- endif %}
{%- set handled_keys = ['name', 'type', 'description', 'enum'] %}
{{- render_extra_keys(param_fields, handled_keys) }}
{{- '\n</parameter>' }}
{%- endfor %}
{%- endif %}
{% set handled_keys = ['type', 'properties', 'required'] %}
{{- render_extra_keys(tool.parameters, handled_keys) }}
{%- if tool.parameters is defined and tool.parameters.required is defined %}
{{- '\n<required>' ~ (tool.parameters.required | tojson) ~ '</required>' }}
{%- endif %}
{{- '\n</parameters>' }}
{%- set handled_keys = ['type', 'name', 'description', 'parameters'] %}
{{- render_extra_keys(tool, handled_keys) }}
{{- '\n</function>' }}
{%- endfor %}
{{- "\n</tools>" }}
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
{%- endif %}
{%- if system_message is defined %}
{{- '<|im_end|>\n' }}
{%- else %}
{%- if tools is iterable and tools | length > 0 %}
{{- '<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- for message in loop_messages %}
{%- if message.role == "assistant" %}
{# Add reasoning content in to content field for unified processing below. #}
{%- if message.reasoning_content is defined and message.reasoning_content is string and message.reasoning_content | trim | length > 0 %}
{%- set content = "<think>\n" ~ message.reasoning_content ~ "\n</think>\n" ~ (message.content | default('', true)) %}
{%- else %}
{%- set content = message.content | default('', true) %}
{%- if content is string -%}
{# Allow downstream logic to to take care of broken thought, only handle coherent reasoning here. #}
{%- if '<think>' not in content and '</think>' not in content -%}
{%- set content = "<think></think>" ~ content -%}
{# Fix content with closing think but not opening #}
{%- elif '<think>' not in content and '</think>' in content -%}
{%- set content = "<think>
" ~ content -%}
{%- endif -%}
{%- else -%}
{%- set content = content -%}
{%- endif -%}
{%- endif %}
{%- if message.tool_calls is defined and message.tool_calls is iterable and message.tool_calls | length > 0 %}
{# Assistant message has tool calls. #}
{{- '<|im_start|>assistant\n' }}
{%- set include_content = not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
{%- if content is string and content | trim | length > 0 %}
{%- if include_content %}
{{- (content | trim) ~ '\n' -}}
{%- else %}
{%- set c = (content | string) %}
{%- if '</think>' in c %}
{# Keep only content after the last closing think. Also generation prompt causes this. #}
{%- set c = (c.split('</think>')|last) %}
{%- elif '<think>' in c %}
{# If <think> was opened but never closed, drop the trailing think segment #}
{%- set c = (c.split('<think>')|first) %}
{%- endif %}
{%- set c = "<think></think>" ~ c | trim %}
{%- if c | length > 0 %}
{{- c ~ '\n' -}}
{%- endif %}
{%- endif %}
{%- else %}
{{- "<think></think>" -}}
{%- endif %}
{%- for tool_call in message.tool_calls %}
{%- if tool_call.function is defined %}
{%- set tool_call = tool_call.function %}
{%- endif %}
{{- '<tool_call>\n<function=' ~ tool_call.name ~ '>\n' -}}
{%- if tool_call.arguments is defined %}{%- if tool_call.arguments is mapping %}
{%- for args_name, args_value in tool_call.arguments|items %}
{{- '<parameter=' ~ args_name ~ '>\n' -}}
{# FIX: Added default('', true) to tool args #}
{%- set args_value = args_value | tojson if args_value is mapping or (args_value is iterable and args_value is not string) else args_value | default('', true) | string %}
{{- args_value ~ '\n</parameter>\n' -}}
{%- endfor %}{%- endif %}
{%- endif %}
{{- '</function>\n</tool_call>\n' -}}
{%- endfor %}
{{- '<|im_end|>\n' }}
{%- else %}
{# Assistant message doesn't have tool calls. #}
{%- if not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
{{- '<|im_start|>assistant\n' ~ (content | default('', true) | string | trim) ~ '<|im_end|>\n' }}
{%- else %}
{%- set c = (content | default('', true) | string) %}
{%- if '<think>' in c and '</think>' in c %}
{%- set c = "<think></think>" ~ (c.split('</think>')|last) %}
{%- endif %}
{%- set c = c | trim %}
{%- if c | length > 0 %}
{{- '<|im_start|>assistant\n' ~ c ~ '<|im_end|>\n' }}
{%- else %}
{{- '<|im_start|>assistant\n<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- endif %}
{%- elif message.role == "user" or message.role == "system" %}
{{- '<|im_start|>' + message.role + '\n' }}
{# FIX: Safely handle null message content #}
{%- set content = message.content | default('', true) | string %}
{{- content }}
{{- '<|im_end|>\n' }}
{%- elif message.role == "tool" %}
{%- if loop.previtem and loop.previtem.role != "tool" %}
{{- '<|im_start|>user\n' }}
{%- endif %}
{{- '<tool_response>\n' }}
{{- message.content | default('', true) }}
{{- '\n</tool_response>\n' }}
{%- if not loop.last and loop.nextitem.role != "tool" %}
{{- '<|im_end|>\n' }}
{%- elif loop.last %}
{{- '<|im_end|>\n' }}
{%- endif %}
{%- else %}
{{- '<|im_start|>' + message.role + '\n' + (message.content | default('', true)) + '<|im_end|>\n' }}
{%- endif %}
{%- endfor %}
{%- if add_generation_prompt %}
{%- if enable_thinking %}
{{- '<|im_start|>assistant\n<think>\n' }}
{%- else %}
{{- '<|im_start|>assistant\n<think></think>' }}
{%- endif %}
{%- endif %}
{# Copyright 2025-present Unsloth. Apache 2.0 License. #}

30
special_tokens_map.json Normal file
View File

@@ -0,0 +1,30 @@
{
"bos_token": {
"content": "<s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false
},
"eos_token": {
"content": "<|im_end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false
},
"pad_token": {
"content": "<SPECIAL_999>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false
},
"unk_token": {
"content": "<unk>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false
}
}

3
tokenizer.json Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:623c34567aebb18582765289fbe23d901c62704d6518d71866e0e58db892b5b7
size 17077484

8021
tokenizer_config.json Normal file
View File

File diff suppressed because one or more lines are too long