INDEX

Explanations

short phrases with the word "a."

oai_token-act-pair · gpt-3.5-turbo

the article "a" frequently used 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ATURES

-0.82

casters

-0.80

iffs

-0.74

stars

-0.72

Americans

-0.70

 Egyptians

-0.69

fn

-0.69

ENTS

-0.69

ometers

-0.68

killers

-0.68

POSITIVE LOGITS

 nutshell

1.23

 manner

1.07

 vein

1.02

 hurry

1.02

 vain

0.99

 guise

0.98

 crowded

0.94

 densely

0.94

 timely

0.89

 midst

0.88

Activations Density 0.173%

short phrases with the word "a."

the article "a" frequently used in various contexts

No Comments

No Known Activations