INDEX

Explanations

causative terms, indicating actions leading to certain consequences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating causal relationships or changes in society

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Moss

-0.66

 Owens

-0.64

urden

-0.63

 moss

-0.61

 Ware

-0.61

 Pool

-0.61

 Winc

-0.60

 Licensed

-0.60

 Monitor

-0.59

Cola

-0.59

POSITIVE LOGITS

 revolutions

0.86

 havoc

0.80

MpServer

0.76

EStream

0.74

riots

0.70

uate

0.70

ocument

0.70

choes

0.69

Discover

0.68

versible

0.67

Activations Density 0.044%

causative terms, indicating actions leading to certain consequences

phrases indicating causal relationships or changes in society

No Comments

No Known Activations