Trelis_-_99-base-checkpoints-v10-gguf – AI Model by RichardErkhov | AlphaNeural AI

You can deploy this model and start earning money today!

RichardErkhov/
Trelis_-_99-base-checkpoints-v10-gguf

like

0

endpoints_compatible

Views

No views yet

Quantization made by Richard Erkhov.

Request more models

99-base-checkpoints-v10 - GGUF

Model creator: https://huggingface.co/Trelis/
Original model: https://huggingface.co/Trelis/99-base-checkpoints-v10/

Name	Quant method	Size
99-base-checkpoints-v10.Q2_K.gguf	Q2_K	0.06GB
99-base-checkpoints-v10.IQ3_XS.gguf	IQ3_XS	0.06GB
99-base-checkpoints-v10.IQ3_S.gguf	IQ3_S	0.06GB
99-base-checkpoints-v10.Q3_K_S.gguf	Q3_K_S	0.06GB
99-base-checkpoints-v10.IQ3_M.gguf	IQ3_M	0.07GB
99-base-checkpoints-v10.Q3_K.gguf	Q3_K	0.07GB
99-base-checkpoints-v10.Q3_K_M.gguf	Q3_K_M	0.07GB
99-base-checkpoints-v10.Q3_K_L.gguf	Q3_K_L	0.07GB
99-base-checkpoints-v10.IQ4_XS.gguf	IQ4_XS	0.07GB
99-base-checkpoints-v10.Q4_0.gguf	Q4_0	0.07GB
99-base-checkpoints-v10.IQ4_NL.gguf	IQ4_NL	0.07GB
99-base-checkpoints-v10.Q4_K_S.gguf	Q4_K_S	0.07GB
99-base-checkpoints-v10.Q4_K.gguf	Q4_K	0.08GB
99-base-checkpoints-v10.Q4_K_M.gguf	Q4_K_M	0.08GB
99-base-checkpoints-v10.Q4_1.gguf	Q4_1	0.07GB
99-base-checkpoints-v10.Q5_0.gguf	Q5_0	0.08GB
99-base-checkpoints-v10.Q5_K_S.gguf	Q5_K_S	0.08GB
99-base-checkpoints-v10.Q5_K.gguf	Q5_K	0.08GB
99-base-checkpoints-v10.Q5_K_M.gguf	Q5_K_M	0.08GB
99-base-checkpoints-v10.Q5_1.gguf	Q5_1	0.08GB
99-base-checkpoints-v10.Q6_K.gguf	Q6_K	0.1GB
99-base-checkpoints-v10.Q8_0.gguf	Q8_0	0.1GB

Original model description:

library_name: transformers base_model: Trelis/SmolLM-135M-layer-pruned-90M-raw tags:

trl
sft
generated_from_trainer model-index:
name: 99-v10-base results: []

99-v10-base

This model is a fine-tuned version of Trelis/SmolLM-135M-layer-pruned-90M-raw on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.9658

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: 0.001
train_batch_size: 16
eval_batch_size: 16
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 4
total_train_batch_size: 512
total_eval_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.01
lr_scheduler_warmup_steps: 19
training_steps: 1951

Training results

Training Loss	Epoch	Step	Validation Loss
0.7232	0.1999	390	0.6950
0.7878	0.3998	780	0.7088
0.5667	0.5997	1170	0.5514
0.5358	0.7996	1560	0.7882
0.4758	0.9995	1950	0.9658

Framework versions

Transformers 4.44.2
Pytorch 2.1.1+cu121
Datasets 3.0.0
Tokenizers 0.19.1