INDEX

Explanations

phrases that connect or link ideas, often represented by the word "and"

oai_token-act-pair · gpt-3.5-turbo

the conjunction "and" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

asper

-0.72

bugs

-0.71

acs

-0.70

FACE

-0.67

"],

-0.66

emi

-0.65

aven

-0.65

asar

-0.65

MRI

-0.64

graded

-0.63

POSITIVE LOGITS

 therefore

1.03

 preferably

1.00

 consequently

1.00

 furthermore

0.97

 moreover

0.93

 indeed

0.91

 possibly

0.89

 hopefully

0.87

 thence

0.87

 perhaps

0.85

Activations Density 0.264%

phrases that connect or link ideas, often represented by the word "and"

the conjunction "and" in various contexts

No Comments

No Known Activations