INDEX

Explanations

phrases related to foundational beliefs or principles, especially related to topics like law, success, and societal structures

oai_token-act-pair · gpt-3.5-turbo

phrases indicating dependence or foundational beliefs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

©¶æ¥µ

-0.78

hai

-0.75

NN

-0.75

Ī

-0.74

VI

-0.73

©¶æ

-0.73

zon

-0.72

ĳå£«

-0.72

heastern

-0.72

vice

-0.72

POSITIVE LOGITS

 belief

0.99

 repetition

0.96

 misunderstanding

0.93

 principles

0.93

 patience

0.92

 conviction

0.91

 adherence

0.91

 necessity

0.91

 conjecture

0.90

 compromises

0.89

Activations Density 0.243%

phrases related to foundational beliefs or principles, especially related to topics like law, success, and societal structures

phrases indicating dependence or foundational beliefs

No Comments

No Known Activations