INDEX

Explanations

references to specific time periods or events mentioned in a document

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the" and phrases indicating time or duration

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

meal

-0.68

arth

-0.67

ragon

-0.67

marine

-0.64

Pad

-0.64

Cho

-0.63

Sax

-0.62

won

-0.61

egu

-0.61

arde

-0.60

POSITIVE LOGITS

 sake

1.45

 purposes

1.19

 reasons

1.01

icion

0.86

 meantime

0.85

 example

0.85

 instance

0.77

 curious

0.73

 Reasons

0.73

 unin

0.71

Activations Density 0.095%

references to specific time periods or events mentioned in a document

occurrences of the word "the" and phrases indicating time or duration

No Comments

No Known Activations