INDEX

Explanations

mentions of the word "resources," especially when emphasizing the availability or allocation of resources

oai_token-act-pair · gpt-3.5-turbo

references to resources and their allocation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

alone

-0.74

ker

-0.73

lihood

-0.73

arella

-0.71

rooms

-0.69

ŃĶ

-0.69

bone

-0.68

 Repe

-0.66

subject

-0.66

emn

-0.66

POSITIVE LOGITS

 resources

1.26

ourcing

0.94

 Resources

0.94

ourced

0.90

 expended

0.79

 allocations

0.77

available

0.76

 allocation

0.76

 resource

0.75

 management

0.75

Activations Density 0.013%

mentions of the word "resources," especially when emphasizing the availability or allocation of resources

references to resources and their allocation

No Comments

No Known Activations