INDEX

Explanations

adjectives ending in 'y'

oai_token-act-pair · gpt-3.5-turbo

the word "thorny" in various contexts related to complex issues

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

 PowerPoint

-0.80

 Examiner

-0.80

orer

-0.68

 Catalog

-0.64

 Annotations

-0.63

 Shooter

-0.63

 Collector

-0.62

 CLSID

-0.62

ickr

-0.62

MAS

-0.60

POSITIVE LOGITS

lly

0.93

yy

0.88

ikes

0.86

ous

0.85

ness

0.83

acies

0.82

wash

0.80

cling

0.80

ielding

0.78

aky

0.78

Activations Density 0.036%

adjectives ending in 'y'

the word "thorny" in various contexts related to complex issues

No Comments

No Known Activations