INDEX

Explanations

phrases related to societal or political issues and controversies

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

aceae

-0.90

igmatic

-0.77

è£

-0.71

lio

-0.70

rique

-0.67

utical

-0.67

sburg

-0.67

llular

-0.65

onder

-0.64

channelAvailability

-0.64

POSITIVE LOGITS

 unfair

0.96

 inaction

0.92

 perceived

0.90

 criticism

0.88

 injustice

0.87

 criticisms

0.87

 accusations

0.86

 questioning

0.83

 suggestions

0.82

 injust

0.81

Activations Density 2.702%

phrases related to societal or political issues and controversies

No Comments

No Known Activations