Get the app
← back to catalog · registered 2026-08-22 13:56

imi2/llama-3-105B-Instruct-abliterated-merged

imi2 Llama 105B
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model. 

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/imi2%2Fllama-3-105B-Instruct-abliterated-merged"
Response includes
  • classification m5
  • files 30
  • hub_downloads_all_time 199
  • author_summary 2 models
  • readme_text full
Try next
GET/models?author=imi2
All 2 models by imi2
 GET/models?family=llama&method=m5
Sibling llama models · same method
 GET/methods/m5
What "m5" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M5
Primary method

Mergekit merge

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
M1 Direct removal →
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • merge tag / mergekit / dare-ties in tags or name
  • no unusual architecture pattern (regular merge)
  • abliterated marker present
How we label → About M5 →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
199
12 last 30d - cooling
Likes
2
Model age
2.4y ago
created 2024-05-07
Downloads over time
Now204→from2↑10,100%
0751502242 on Jul 24, 2024204 on Oct 11204 on Oct 10Jul '24Nov '24Mar '25Jul '25Nov '25MarJul
Jul 24, 2024 → Oct 11 · 155 snapshots · spans 809 days

Metadata

Tags
transformers safetensors llama text-generation mergekit merge conversational text-generation-inference endpoints_compatible region:us

Related

Other by imi2
llama-3-120B-Instruct-abliterated-merged 20
Total size
195 GB
Files
30
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2024-05-08 05:43
Source huggingface.co/imi2/llama-3-105B-Instruct-abliterated-merged ↗

Files by quantization

Auxiliary files 30 files 195 GB
model-00013-of-00022.safetensors 9.31 GB 8304e2cd download
model-00018-of-00022.safetensors 9.28 GB 446ddf18 download
model-00017-of-00022.safetensors 9.28 GB 0a05a93b download
model-00007-of-00022.safetensors 9.25 GB 912c5bea download
model-00012-of-00022.safetensors 9.25 GB 4426bb1e download
model-00010-of-00022.safetensors 9.25 GB cbd0b7c9 download
model-00015-of-00022.safetensors 9.25 GB 33830d10 download
model-00006-of-00022.safetensors 9.25 GB 2696bbcb download
model-00016-of-00022.safetensors 9.25 GB 99e407e2 download
model-00001-of-00022.safetensors 9.21 GB e0f089b8 download
model-00021-of-00022.safetensors 9.16 GB 511af76e download
model-00004-of-00022.safetensors 9.16 GB aba1e922 download
model-00002-of-00022.safetensors 9.13 GB 77f908ab download
model-00019-of-00022.safetensors 9.13 GB b7452de8 download
model-00011-of-00022.safetensors 9.13 GB 30011ea8 download
model-00008-of-00022.safetensors 9.00 GB e56f0b40 download
model-00009-of-00022.safetensors 9.00 GB 0a50540e download
model-00014-of-00022.safetensors 9.00 GB 65774be2 download
model-00005-of-00022.safetensors 9.00 GB b01bb863 download
model-00020-of-00022.safetensors 8.97 GB db5b6080 download
model-00003-of-00022.safetensors 8.97 GB 472a0965 download
model-00022-of-00022.safetensors 2.96 GB 4cce2c08 download
tokenizer.json 8.66 MB b197f72e download
model.safetensors.index.json 83.2 KB 3fed6048 download
tokenizer_config.json 49.8 KB 870479e3 download
.gitattributes 1.48 KB a6344aac download
README.md 868 B 40002dca download
config.json 726 B 28fdbd79 download
mergekit_config.yml 357 B a84471f6 download
special_tokens_map.json 301 B cfabacc2 download

README current version from Hugging Face

base_model: []
library_name: transformers
tags:

  • mergekit
  • merge

105B

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

  • ../../Storage/failspy_llama-3-70B-Instruct-abliterated

Configuration

The following YAML configuration was used to produce this model:

dtype: float16
merge_method: passthrough
slices:
- sources:
  - layer_range: [0, 40]
    model: ../../Storage/failspy_llama-3-70B-Instruct-abliterated
- sources:
  - layer_range: [20, 60]
    model: ../../Storage/failspy_llama-3-70B-Instruct-abliterated
- sources:
  - layer_range: [40, 80]
    model: ../../Storage/failspy_llama-3-70B-Instruct-abliterated

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2024-05-08Upload folder using huggingface_hub (#1)9ab039a868 B
    Loading...

Discussions 1 thread

  1. 2024-05-08PRUpload folder using huggingface_hubmerged1 💬#1
    Loading...

Continue on Hugging Face →

Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration