INDEX

Explanations

numbers associated with measurements, specifically in milligrams

oai_token-act-pair · gpt-3.5-turbo

references to milligrams in various contexts

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

 pupil

-0.75

gotten

-0.73

KER

-0.70

*/(

-0.70

 Emir

-0.66

BMC

-0.66

Catal

-0.64

ļéĨĴ

-0.64

 Horizons

-0.64

Saint

-0.63

POSITIVE LOGITS

mg

1.29

mt

1.14

omez

0.96

emonic

0.91

ertodd

0.90

achev

0.87

ml

0.86

ombie

0.85

irez

0.85

ilit

0.84

Activations Density 0.005%

numbers associated with measurements, specifically in milligrams

references to milligrams in various contexts

No Comments

No Known Activations