INDEX

Explanations

references to the concept of 'back'

oai_token-act-pair · gpt-3.5-turbo

the term "Back" or variations of it 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

course

-0.71

 appointment

-0.66

ominated

-0.66

 magistrate

-0.65

icable

-0.64

ECA

-0.64

chance

-0.62

lich

-0.62

ich

-0.60

adm

-0.59

POSITIVE LOGITS

 Back

3.44

Back

2.53

 BACK

2.29

back

1.90

 back

1.83

BACK

1.74

 backs

1.57

backs

1.49

 Forward

1.24

 Front

1.21

Activations Density 0.012%

references to the concept of 'back'

the term "Back" or variations of it in various contexts

No Comments

No Known Activations