INDEX

Explanations

terms that are related to artistic qualities or descriptions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lain

-0.86

leaf

-0.85

ãģ®éŃĶ

-0.84

clips

-0.81

ortium

-0.78

bourg

-0.77

foreseen

-0.77

oval

-0.76

handle

-0.76

FACE

-0.72

POSITIVE LOGITS

 tendencies

1.05

 persuasion

0.85

 impulses

0.80

 fallacy

0.79

 motives

0.79

 worldview

0.76

 approach

0.73

 interpretation

0.73

 mindset

0.71

 outlook

0.71

Activations Density 0.014%

terms that are related to artistic qualities or descriptions

No Comments

No Known Activations