INDEX

Explanations

phrases related to significant events or developments

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a list or items being presented

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

nels

-0.73

iffe

-0.72

rx

-0.68

ede

-0.68

each

-0.67

arry

-0.66

 relies

-0.66

Measure

-0.65

 emphasizes

-0.64

gyn

-0.64

POSITIVE LOGITS

 obligatory

1.25

 beginnings

1.24

 usual

1.16

 slightest

1.13

 culmination

1.12

 dreaded

1.10

 finest

1.09

 definitive

1.07

 simplest

1.07

 smallest

1.06

Activations Density 0.514%

phrases related to significant events or developments

phrases that indicate a list or items being presented

No Comments

No Known Activations