llama-coc-sentiment

maoelana/llama-coc-sentiment

llama-coc-sentiment is a machine learning model from maoelana. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.

- Developed by: maoelana - License: apache-2.0 - Finetuned from model : unsloth/llama-3-8b-bnb-4bit

Downloads · 30 days

Access

Public

Updated Jul 13, 2024

Repo size

168 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors168 MB · 100%

At a glance

Library: transformers
License: apache-2.0
Access: Public
Created: Jul 13, 2024
Updated: Jul 13, 2024
SHA: 48339651

Base models

unsloth/llama-3-8b-bnb-4bit

Model card

From the Hugging Face model README

Uploaded model

Developed by: maoelana
License: apache-2.0
Finetuned from model : unsloth/llama-3-8b-bnb-4bit

Step	Epoch	Learning Rate	Loss
1	0.0005491488193300384	4e-05	3.3219
2	0.001098297638660077	8e-05	3.2933
3	0.0016474464579901153	0.00012	3.2393
4	0.002196595277320154	0.00016	3.1684
5	0.0027457440966501922	0.0002	2.34
6	0.0032948929159802307	0.00019636363636363636	2.0505
7	0.003844041735310269	0.00019272727272727274	1.7822
8	0.004393190554640308	0.0001890909090909091	1.3199
9	0.004942339373970346	0.00018545454545454545	0.972
10	0.0054914881933003845	0.00018181818181818183	0.6681
11	0.0060406370126304225	0.0001781818181818182	1.1239
12	0.006589785831960461	0.00017454545454545454	0.8348
13	0.007138934651290499	0.0001709090909090909	0.6586
14	0.007688083470620538	0.00016727272727272728	0.8314
15	0.008237232289950576	0.00016363636363636366	0.6409
16	0.008786381109280615	0.00016	0.664
17	0.009335529928610654	0.00015636363636363637	0.9784
18	0.009884678747940691	0.00015272727272727275	1.0465
19	0.01043382756727073	0.0001490909090909091	0.7074
20	0.010982976386600769	0.00014545454545454546	1.4477
21	0.011532125205930808	0.00014181818181818184	1.3083
22	0.012081274025260845	0.0001381818181818182	0.3509
23	0.012630422844590884	0.00013454545454545455	0.4807
24	0.013179571663920923	0.00013090909090909093	0.6549
25	0.013728720483250962	0.00012727272727272728	0.7044
26	0.014277869302580999	0.00012363636363636364	0.8933
27	0.014827018121911038	0.00012	0.6275
28	0.015376166941241077	0.00011636363636363636	0.4381
29	0.015925315760571115	0.00011272727272727272	0.7067
30	0.016474464579901153	0.00010909090909090909	0.7428
31	0.017023613399231193	0.00010545454545454545	0.9587
32	0.01757276221856123	0.00010181818181818181	1.0544
33	0.018121911037891267	9.818181818181818e-05	0.5977
34	0.018671059857221308	9.454545454545455e-05	0.8479
35	0.019220208676551345	9.090909090909092e-05	0.6241
36	0.019769357495881382	8.727272727272727e-05	0.6209
37	0.020318506315211423	8.363636363636364e-05	1.0284
38	0.02086765513454146	8e-05	0.5394
39	0.0214168039538715	7.636363636363637e-05	0.6672
40	0.021965952773201538	7.272727272727273e-05	0.1762
41	0.022515101592531575	6.90909090909091e-05	0.4435
42	0.023064250411861616	6.545454545454546e-05	0.5048
43	0.023613399231191653	6.181818181818182e-05	0.4497
44	0.02416254805052169	5.818181818181818e-05	0.9188
45	0.02471169686985173	5.4545454545454546e-05	0.4881
46	0.025260845689181768	5.090909090909091e-05	0.3593
47	0.02580999450851181	4.7272727272727275e-05	0.8929
48	0.026359143327841845	4.3636363636363636e-05	0.6681
49	0.026908292147171883	4e-05	0.7153
50	0.027457440966501923	3.6363636363636364e-05	0.5856
51	0.02800658978583196	3.272727272727273e-05	1.0362
52	0.028555738605161998	2.909090909090909e-05	0.546
53	0.029104887424492038	2.5454545454545454e-05	0.9543
54	0.029654036243822075	2.1818181818181818e-05	0.988
55	0.030203185063152116	1.8181818181818182e-05	0.3717
56	0.030752333882482153	1.4545454545454545e-05	0.9083
57	0.03130148270181219	1.0909090909090909e-05	0.8964
58	0.03185063152114223	7.272727272727272e-06	0.6579
59	0.032399780340472265	3.636363636363636e-06	0.3562
60	0.032948929159802305	0.0	0.3829

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

Library: transformers
License: apache-2.0
Languages: en
Created: Jul 13, 2024
Updated: Jul 13, 2024