Lambent / CosMoE-Lisa-4x1b

huggingface.co
Total runs: 11
24-hour runs: 0
7-day runs: -2
30-day runs: 2
Model's Last Updated: June 30 2025
text-generation

Introduction of CosMoE-Lisa-4x1b

Model Details of CosMoE-Lisa-4x1b

Model AGIEval GPT4All TruthfulQA Bigbench Average
CosMoE-Lisa-4x1b 23.42 43.56 38.11 28.35 33.36

Ambitious training, but appears to have decreased capabilities rather than aided them. Lessons learned.

Built with Axolotl

See axolotl config

axolotl version: 0.4.0

base_model: Lambent/cosmoem-4x1b
model_type: AutoModelForCausalLM
tokenizer_type: LlamaTokenizer
trust_remote_code: true

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: Vezora/Tested-22k-Python-Alpaca
    type: alpaca
  - path: teknium/GPTeacher-General-Instruct #89.3k
    type: gpteacher
  - path: HuggingFaceTB/cosmopedia-100k
    type: completion
dataset_prepared_path: prepared-education
val_set_size: 0.05
output_dir: ./lisa-out

sequence_len: 2048
sample_packing: true
eval_sample_packing: false
pad_to_sequence_len: true

lisa_n_layers: 4
lisa_step_interval: 10
lisa_layers_attribute: model.layers

adapter:
lora_model_dir:
lora_r:
lora_alpha:
lora_dropout:
lora_target_linear:
lora_fan_in_fan_out:

wandb_project: CosMoE-Lisa-4x1b
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 1
micro_batch_size: 1
num_epochs: 2
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0005

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

loss_watchdog_threshold: 5.0
loss_watchdog_patience: 3

warmup_steps: 20
evals_per_epoch: 2
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.002
fsdp:
fsdp_config:
special_tokens:

lisa-out

This model is a fine-tuned version of Lambent/cosmoem-4x1b on the None dataset. It achieves the following results on the evaluation set:

  • Loss: nan
Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure
Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0005
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 20
  • num_epochs: 2
Training results
Training Loss Epoch Step Validation Loss
0.8891 0.0 1 nan
2.1984 0.5 30501 nan
1.5296 1.0 61002 nan
1.4647 1.49 91503 nan
1.4698 1.99 122004 nan
Framework versions
  • Transformers 4.40.0.dev0
  • Pytorch 2.1.2+cu118
  • Datasets 2.18.0
  • Tokenizers 0.15.0

Runs of Lambent CosMoE-Lisa-4x1b on huggingface.co

11
Total runs
0
24-hour runs
-3
3-day runs
-2
7-day runs
2
30-day runs

More Information About CosMoE-Lisa-4x1b huggingface.co Model

More CosMoE-Lisa-4x1b license Visit here:

https://choosealicense.com/licenses/apache-2.0

CosMoE-Lisa-4x1b huggingface.co

CosMoE-Lisa-4x1b huggingface.co is an AI model on huggingface.co that provides CosMoE-Lisa-4x1b's model effect (), which can be used instantly with this Lambent CosMoE-Lisa-4x1b model. huggingface.co supports a free trial of the CosMoE-Lisa-4x1b model, and also provides paid use of the CosMoE-Lisa-4x1b. Support call CosMoE-Lisa-4x1b model through api, including Node.js, Python, http.

CosMoE-Lisa-4x1b huggingface.co Url

https://huggingface.co/Lambent/CosMoE-Lisa-4x1b

Lambent CosMoE-Lisa-4x1b online free

CosMoE-Lisa-4x1b huggingface.co is an online trial and call api platform, which integrates CosMoE-Lisa-4x1b's modeling effects, including api services, and provides a free online trial of CosMoE-Lisa-4x1b, you can try CosMoE-Lisa-4x1b online for free by clicking the link below.

Lambent CosMoE-Lisa-4x1b online free url in huggingface.co:

https://huggingface.co/Lambent/CosMoE-Lisa-4x1b

CosMoE-Lisa-4x1b install

CosMoE-Lisa-4x1b is an open source model from GitHub that offers a free installation service, and any user can find CosMoE-Lisa-4x1b on GitHub to install. At the same time, huggingface.co provides the effect of CosMoE-Lisa-4x1b install, users can directly use CosMoE-Lisa-4x1b installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

CosMoE-Lisa-4x1b install url in huggingface.co:

https://huggingface.co/Lambent/CosMoE-Lisa-4x1b

Url of CosMoE-Lisa-4x1b

CosMoE-Lisa-4x1b huggingface.co Url

Provider of CosMoE-Lisa-4x1b huggingface.co

Lambent
ORGANIZATIONS

Other API from Lambent

huggingface.co

Total runs: 109
Run Growth: 42
Growth Rate: 38.53%
Updated:March 16 2024
huggingface.co

Total runs: 40
Run Growth: 18
Growth Rate: 45.00%
Updated:March 17 2026
huggingface.co

Total runs: 26
Run Growth: 12
Growth Rate: 46.15%
Updated:March 26 2026
huggingface.co

Total runs: 12
Run Growth: -16
Growth Rate: -133.33%
Updated:March 09 2026