Sponsored by APIMaster
Toolify Toolify logo
AI Models
hrm-text-tiny-random

yujiepan / hrm-text-tiny-random

Run on huggingface.co
huggingface.co
Total runs: 312
24-hour runs: 0
7-day runs: 159
30-day runs: 251
Model's Last Updated: May 27 2026
text-generation

Introduction of hrm-text-tiny-random

Model Details of 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

Runs of yujiepan hrm-text-tiny-random on huggingface.co

312
Total runs
0
24-hour runs
-2
3-day runs
159
7-day runs
251
30-day runs

More Information About hrm-text-tiny-random huggingface.co Model

hrm-text-tiny-random huggingface.co

hrm-text-tiny-random huggingface.co is an AI model on huggingface.co that provides hrm-text-tiny-random's model effect (), which can be used instantly with this yujiepan hrm-text-tiny-random model. huggingface.co supports a free trial of the hrm-text-tiny-random model, and also provides paid use of the hrm-text-tiny-random. Support call hrm-text-tiny-random model through api, including Node.js, Python, http.

hrm-text-tiny-random huggingface.co Url

https://huggingface.co/yujiepan/hrm-text-tiny-random

yujiepan hrm-text-tiny-random online free

hrm-text-tiny-random huggingface.co is an online trial and call api platform, which integrates hrm-text-tiny-random's modeling effects, including api services, and provides a free online trial of hrm-text-tiny-random, you can try hrm-text-tiny-random online for free by clicking the link below.

yujiepan hrm-text-tiny-random online free url in huggingface.co:

https://huggingface.co/yujiepan/hrm-text-tiny-random

hrm-text-tiny-random install

hrm-text-tiny-random is an open source model from GitHub that offers a free installation service, and any user can find hrm-text-tiny-random on GitHub to install. At the same time, huggingface.co provides the effect of hrm-text-tiny-random install, users can directly use hrm-text-tiny-random installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

hrm-text-tiny-random install url in huggingface.co:

https://huggingface.co/yujiepan/hrm-text-tiny-random

Url of hrm-text-tiny-random

hrm-text-tiny-random huggingface.co Url

Provider of hrm-text-tiny-random huggingface.co

yujiepan
ORGANIZATIONS
huggingface.co

Other API from yujiepan