INDEX

Explanations

phrases related to conflicts between protestors and law enforcement

oai_token-act-pair · gpt-3.5-turbo

references to police and their involvement in protests or riots

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

vironment

-0.86

igible

-0.78

endon

-0.76

 Tasman

-0.75

ochond

-0.74

Mars

-0.71

 aloud

-0.71

âĶĢâĶĢâĶĢâĶĢ

-0.69

 Lear

-0.68

estead

-0.67

POSITIVE LOGITS

 officers

0.97

 crackdown

0.96

 barric

0.93

 brutality

0.92

 checkpoints

0.90

 detained

0.89

 repression

0.88

 imposed

0.87

 clashed

0.86

 policemen

0.86

Activations Density 0.116%

phrases related to conflicts between protestors and law enforcement

references to police and their involvement in protests or riots

No Comments

No Known Activations