INDEX

Explanations

phrases expressing opinions or judgments

oai_token-act-pair · gpt-3.5-turbo

expressions that indicate perception or opinion regarding something being surprising, odd, or subjective

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

anwhile

-0.85

 Dise

-0.71

cised

-0.70

"]=>

-0.68

etheless

-0.66

rench

-0.61

aughtered

-0.61

ugs

-0.61

 WATCH

-0.61

formance

-0.59

POSITIVE LOGITS

 daunting

1.21

 innocuous

1.11

 trivial

1.04

 intimidating

1.03

 insignificant

1.02

 intuitive

0.95

 esoteric

0.95

 obvious

0.92

 glamorous

0.90

 confusing

0.88

Activations Density 0.140%

phrases expressing opinions or judgments

expressions that indicate perception or opinion regarding something being surprising, odd, or subjective

No Comments

No Known Activations