yujiepan/hrm-text-tiny-random

This tiny model is intended for debugging. It is randomly initialized using the configuration adapted from sapientinc/HRM-Text-1B.

File path	Size
model.safetensors	2.3MB

Example usage:

from transformers import pipeline

model_id = "yujiepan/hrm-text-tiny-random"
pipe = pipeline(
    "text-generation", model=model_id, device="cuda",
    trust_remote_code=True, max_new_tokens=16,
)
print(pipe("Hello World!"))

Codes to create this repo:

Click to expand

import json

import torch

from huggingface_hub import file_exists, hf_hub_download
from transformers import (
    AutoConfig,
    AutoModelForCausalLM,
    AutoTokenizer,
    GenerationConfig,
    pipeline,
    set_seed,
)

source_model_id = "sapientinc/HRM-Text-1B"
save_folder = "/tmp/yujiepan/hrm-text-tiny-random"
tokenizer = AutoTokenizer.from_pretrained(
    source_model_id, trust_remote_code=True,
)
tokenizer.save_pretrained(save_folder)

with open(hf_hub_download(source_model_id, filename='config.json', repo_type='model'), 'r', encoding='utf-8') as f:
    config_json: dict = json.load(f)
config_json.update({
    "hidden_size": 8,
    "intermediate_size": 64,
    "num_attention_heads": 4,
    "num_key_value_heads": 4,
    "head_dim": 32,
    "num_hidden_layers": 8,
})
with open(f"{save_folder}/config.json", "w", encoding='utf-8') as f:
    json.dump(config_json, f, indent=2)

config = AutoConfig.from_pretrained(
    save_folder,
    trust_remote_code=True,
)

model = AutoModelForCausalLM.from_config(
    config,
    dtype=torch.bfloat16,
    trust_remote_code=True,
)
if file_exists(filename="generation_config.json", repo_id=source_model_id, repo_type='model'):
    model.generation_config = GenerationConfig.from_pretrained(
        source_model_id, trust_remote_code=True,
    )
set_seed(42)
model = model.cpu()
with torch.no_grad():
    for name, p in sorted(model.named_parameters()):
        torch.nn.init.normal_(p, 0, 0.2)
        print(name, p.shape)
model.save_pretrained(save_folder)

Printing the model:

Click to expand

HrmTextForCausalLM(
  (model): HrmTextModel(
    (embed_tokens): Embedding(65536, 8, padding_idx=5)
    (rotary_emb): HrmTextRotaryEmbedding()
    (L_module): HrmTextStack(
      (layers): ModuleList(
        (0-7): 8 x HrmTextDecoderLayer(
          (self_attn): HrmTextAttention(
            (q_proj): Linear(in_features=8, out_features=128, bias=False)
            (k_proj): Linear(in_features=8, out_features=128, bias=False)
            (v_proj): Linear(in_features=8, out_features=128, bias=False)
            (o_proj): Linear(in_features=128, out_features=8, bias=False)
            (gate_proj): Linear(in_features=8, out_features=128, bias=False)
          )
          (mlp): HrmTextMLP(
            (gate_proj): Linear(in_features=8, out_features=64, bias=False)
            (up_proj): Linear(in_features=8, out_features=64, bias=False)
            (down_proj): Linear(in_features=64, out_features=8, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): HrmTextRMSNorm(eps=1e-06)
          (post_attention_layernorm): HrmTextRMSNorm(eps=1e-06)
        )
      )
      (final_norm): HrmTextRMSNorm(eps=1e-06)
    )
    (H_module): HrmTextStack(
      (layers): ModuleList(
        (0-7): 8 x HrmTextDecoderLayer(
          (self_attn): HrmTextAttention(
            (q_proj): Linear(in_features=8, out_features=128, bias=False)
            (k_proj): Linear(in_features=8, out_features=128, bias=False)
            (v_proj): Linear(in_features=8, out_features=128, bias=False)
            (o_proj): Linear(in_features=128, out_features=8, bias=False)
            (gate_proj): Linear(in_features=8, out_features=128, bias=False)
          )
          (mlp): HrmTextMLP(
            (gate_proj): Linear(in_features=8, out_features=64, bias=False)
            (up_proj): Linear(in_features=8, out_features=64, bias=False)
            (down_proj): Linear(in_features=64, out_features=8, bias=False)
            (act_fn): SiLUActivation()
          )
          (input_layernorm): HrmTextRMSNorm(eps=1e-06)
          (post_attention_layernorm): HrmTextRMSNorm(eps=1e-06)
        )
      )
      (final_norm): HrmTextRMSNorm(eps=1e-06)
    )
  )
  (lm_head): Linear(in_features=8, out_features=65536, bias=False)
)

Test environment:

• torch: 2.10.0+cu128
• transformers: 5.9.0