Skip to page content
SAVRN
Search Plan a project

Open-weight model

Llama-3.2-1B-vrank-vs-attn

by Sehyun Park sehyun734/Llama-3.2-1B-vrank-vs-attn

Llama-3.2-1B-vrank-vs-attn is an open-weight model from Sehyun Park. Its published files total 215.8 MB.

Parameters—
Context—
Weights215.3 MB
License—
AccessOpen weights
Monthly Downloads—

Model Card

The publisher has not written a card for this model.

Identity and Version

Repository
sehyun734/Llama-3.2-1B-vrank-vs-attn
Publisher
Sehyun Park
Task
Not stated by the source
Modality
Other
Library
Not stated by the source
Parameters
Not stated by the source
Languages
Not stated by the source
Revision
2b64a966a260f12b32d2c502e8b5389dee467517
First published
2026-09-29
Last updated
2026-10-02

Files and Weights

51 files, 215.8 MB in total. The weights are 19 files totalling 215.3 MB in safetensors.

Weights19 files · 215.3 MB
Configuration31 files · 536.1 KB
Repository1 file · 1.5 KB
Every file
FileTypeSizeSHA-256
lora/rank=16,learning_rate=0.0001,seed=0/lora.safetensorsWeights9.4 MB c3897a690afb
lora/rank=16,learning_rate=0.0003,seed=0/lora.safetensorsWeights9.4 MB f27a88623ddb
lora/rank=16,learning_rate=0.001,seed=0/lora.safetensorsWeights9.4 MB a5e593360df7
lora/rank=16,learning_rate=0.003,seed=0/lora.safetensorsWeights9.4 MB 4d11bc105b51
lora/rank=4,learning_rate=0.0001,seed=0/lora.safetensorsWeights2.4 MB ca472e0367d0
lora/rank=4,learning_rate=0.0003,seed=0/lora.safetensorsWeights2.4 MB dead38f725a4
lora/rank=4,learning_rate=0.001,seed=0/lora.safetensorsWeights2.4 MB 4de2fb48cfe5
lora/rank=4,learning_rate=0.003,seed=0/lora.safetensorsWeights2.4 MB 2e358a9a2f95
lora/rank=64,learning_rate=0.0001,seed=0/lora.safetensorsWeights37.8 MB 47215a1d7a93
lora/rank=64,learning_rate=0.0003,seed=0/lora.safetensorsWeights37.8 MB 3351d4be5529
lora/rank=64,learning_rate=0.001,seed=0/lora.safetensorsWeights37.8 MB d4aa9427e8bc
lora/rank=64,learning_rate=0.003,seed=0/lora.safetensorsWeights37.8 MB 37fd4cd40f0e
prefix/prefix_len=16,learning_rate=0.0001,seed=0/prefix.safetensorsWeights1.1 MB caf0116557d7
prefix/prefix_len=16,learning_rate=0.0003,seed=0/prefix.safetensorsWeights1.1 MB 8fc34359c2bc
prefix/prefix_len=16,learning_rate=0.001,seed=0/prefix.safetensorsWeights1.1 MB b70d7bd448da
prefix/prefix_len=16,learning_rate=0.003,seed=0/prefix.safetensorsWeights1.1 MB d657abef9a9d
prefix/prefix_len=64,learning_rate=0.0001,seed=0/prefix.safetensorsWeights4.2 MB 4b6c56755950
prefix/prefix_len=64,learning_rate=0.0003,seed=0/prefix.safetensorsWeights4.2 MB 9e4ad974f882
prefix/prefix_len=64,learning_rate=0.001,seed=0/prefix.safetensorsWeights4.2 MB 475e420e1d46
base/attn.jsonConfiguration32.6 KB —
base/gsm8k.jsonConfiguration38 B —
base/vrank.jsonConfiguration6.1 KB —
lora/rank=16,learning_rate=0.0001,seed=0/attn.jsonConfiguration41.0 KB —
lora/rank=16,learning_rate=0.0001,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=16,learning_rate=0.0001,seed=0/vrank.jsonConfiguration6.1 KB —
lora/rank=16,learning_rate=0.0003,seed=0/attn.jsonConfiguration41.0 KB —
lora/rank=16,learning_rate=0.0003,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=16,learning_rate=0.001,seed=0/attn.jsonConfiguration40.9 KB —
lora/rank=16,learning_rate=0.001,seed=0/gsm8k.jsonConfiguration36 B —
lora/rank=16,learning_rate=0.003,seed=0/attn.jsonConfiguration40.7 KB —
lora/rank=16,learning_rate=0.003,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=4,learning_rate=0.0001,seed=0/attn.jsonConfiguration41.1 KB —
lora/rank=4,learning_rate=0.0001,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=4,learning_rate=0.0003,seed=0/attn.jsonConfiguration40.9 KB —
lora/rank=4,learning_rate=0.0003,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=4,learning_rate=0.001,seed=0/attn.jsonConfiguration40.9 KB —
lora/rank=4,learning_rate=0.001,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=4,learning_rate=0.003,seed=0/attn.jsonConfiguration40.8 KB —
lora/rank=4,learning_rate=0.003,seed=0/gsm8k.jsonConfiguration36 B —
lora/rank=64,learning_rate=0.0001,seed=0/attn.jsonConfiguration40.9 KB —
lora/rank=64,learning_rate=0.0001,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=64,learning_rate=0.0003,seed=0/attn.jsonConfiguration40.9 KB —
lora/rank=64,learning_rate=0.0003,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=64,learning_rate=0.001,seed=0/attn.jsonConfiguration40.9 KB —
lora/rank=64,learning_rate=0.001,seed=0/gsm8k.jsonConfiguration37 B —
lora/rank=64,learning_rate=0.003,seed=0/attn.jsonConfiguration40.7 KB —
lora/rank=64,learning_rate=0.003,seed=0/gsm8k.jsonConfiguration38 B —
prefix/prefix_len=16,learning_rate=0.0001,seed=0/gsm8k.jsonConfiguration37 B —
prefix/prefix_len=16,learning_rate=0.0003,seed=0/gsm8k.jsonConfiguration37 B —
prefix/prefix_len=16,learning_rate=0.001,seed=0/gsm8k.jsonConfiguration36 B —
.gitattributesRepository1.5 KB —

License and Download

License
Not stated by the source
Access
Open weights, no gate
Download size
215.3 MB
Download from Sehyun Park

Released by Sehyun Park through its official repository on Hugging Face.

Memory Requirements

PrecisionWeights in memory
As published215.3 MB

Weights only, from the published parameter count; the key-value cache and runtime add to this.