INDEX

Explanations

references to social or political unrest and discontent

oai_token-act-pair · gpt-3.5-turbo

occurrences of social unrest and associated sentiments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

 TAMADRA

-0.70

 Exam

-0.64

 Trees

-0.62

Companies

-0.62

 Lessons

-0.61

 Oral

-0.61

 Distance

-0.61

missions

-0.61

 Fitness

-0.60

 Gifts

-0.60

POSITIVE LOGITS

rained

1.63

 unrest

0.98

raint

0.91

rict

0.84

ief

0.80

orious

0.79

atics

0.77

ful

0.75

fully

0.74

ricting

0.74

Activations Density 0.011%

references to social or political unrest and discontent

occurrences of social unrest and associated sentiments

No Comments

No Known Activations