INDEX

Explanations

information related to incidents or events involving arrests, investigations, crimes, tragedies, and legal proceedings

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aez

-0.91

tick

-0.77

achel

-0.76

Mask

-0.73

DragonMagazine

-0.71

cin

-0.70

eous

-0.70

voy

-0.69

Near

-0.69

ace

-0.68

POSITIVE LOGITS

 completing

0.87

 discovering

0.84

 sunset

0.81

noon

0.81

 failing

0.80

 midnight

0.80

 inspecting

0.80

 receiving

0.80

 sustaining

0.79

words

0.78

Activations Density 10.774%

information related to incidents or events involving arrests, investigations, crimes, tragedies, and legal proceedings

No Comments

No Known Activations