INDEX

Explanations

phrases related to potential or possibility

oai_token-act-pair · gpt-3.5-turbo

references to potential situations or events

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

masters

-0.96

cipline

-0.83

rike

-0.83

erie

-0.82

bane

-0.82

gall

-0.81

gar

-0.80

men

-0.80

worn

-0.80

rix

-0.80

POSITIVE LOGITS

 inclusion

0.81

 Parenthood

0.81

 defect

0.79

 embodiments

0.78

 inference

0.78

 obstruction

0.76

 outcomes

0.76

 explanations

0.75

 guiActiveUn

0.74

 demise

0.73

Activations Density 0.025%

phrases related to potential or possibility

references to potential situations or events

No Comments

No Known Activations