INDEX

Explanations

verbs related to creating and interacting within digital platforms

oai_token-act-pair · gpt-3.5-turbo

instances of designing, building, or constructing activities and initiatives

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

catentry

-0.82

ãĥ¯

-0.64

hra

-0.57

ansion

-0.57

 presided

-0.57

********************************

-0.54

 Ramos

-0.54

UDP

-0.54

 visc

-0.53

 Anat

-0.53

POSITIVE LOGITS

 themselves

0.94

 THEIR

0.90

 their

0.89

their

0.87

 anonymously

0.84

 responsibly

0.78

 freely

0.78

acebook

0.75

 lifestyles

0.71

selves

0.67

Activations Density 0.752%

verbs related to creating and interacting within digital platforms

instances of designing, building, or constructing activities and initiatives

No Comments

No Known Activations