INDEX

Explanations

dates or deadlines mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

references to upcoming events and their timing

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

 Entered

-0.70

apsed

-0.67

 Quantity

-0.66

iuses

-0.62

Avg

-0.62

stitial

-0.61

 Deaths

-0.60

yrics

-0.60

 Example

-0.59

 Errors

-0.59

POSITIVE LOGITS

 tomorrow

1.39

1.19

 someday

1.16

 next

1.06

 upcoming

1.06

 future

1.05

morrow

1.00

 hereafter

1.00

0.99

 forthcoming

0.96

Activations Density 1.878%

dates or deadlines mentioned in a text

references to upcoming events and their timing

No Comments

No Known Activations