INDEX

Explanations

phrases related to celebrating holidays, particularly Christmas and Yule

oai_token-act-pair · gpt-3.5-turbo

words and terms related to "rules" or "regulations."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ITNESS

-0.75

 GOODMAN

-0.75

arella

-0.72

POL

-0.71

enegger

-0.71

ISON

-0.67

chrom

-0.65

Extreme

-0.64

é¾

-0.64

alez

-0.63

POSITIVE LOGITS

ule

0.99

cules

0.89

lette

0.88

ploy

0.87

bum

0.86

lement

0.81

leted

0.76

gments

0.76

ttes

0.75

uge

0.75

Activations Density 0.006%

phrases related to celebrating holidays, particularly Christmas and Yule

words and terms related to "rules" or "regulations."

No Comments

No Known Activations