INDEX

Explanations

words related to humanitarian aid, refugee crises, and shelter

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

agall

-0.76

 quizz

-0.71

 antioxid

-0.71

 Collider

-0.69

ellectual

-0.68

 sophistic

-0.67

hibition

-0.64

Hof

-0.62

 antitrust

-0.61

ouver

-0.61

POSITIVE LOGITS

 Refugees

1.14

ilitating

1.07

 resettlement

0.97

 refugees

0.97

 shelter

0.96

ilitation

0.93

 Refugee

0.92

 shelters

0.91

 Shelter

0.91

 orphans

0.90

Activations Density 1.697%

words related to humanitarian aid, refugee crises, and shelter

No Comments

No Known Activations