INDEX

Explanations

references to the prefix "mon-" followed by a word

oai_token-act-pair · gpt-3.5-turbo

terms related to monopolies and monopolistic behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

EXP

-0.80

ULTS

-0.79

ORN

-0.74

NRS

-0.73

éĹĺ

-0.73

FUL

-0.73

IRO

-0.72

OHN

-0.71

Enhanced

-0.70

 Norn

-0.70

POSITIVE LOGITS

stros

1.08

oton

1.02

olithic

1.01

itored

0.99

olith

0.98

soon

0.96

 gobl

0.95

ologue

0.94

roe

0.90

astical

0.90

Activations Density 0.008%

references to the prefix "mon-" followed by a word

terms related to monopolies and monopolistic behavior

No Comments

No Known Activations