INDEX

Explanations

references to the concept of profiling, especially in the context of racial profiling and computer programming

oai_token-act-pair · gpt-3.5-turbo

terms related to profiling 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

vette

-0.78

pite

-0.75

ija

-0.73

ership

-0.72

win

-0.72

mp

-0.71

birth

-0.70

Interstitial

-0.70

fal

-0.68

Original

-0.66

POSITIVE LOGITS

 profiling

1.43

 profiles

0.99

 prof

0.89

eatures

0.79

ank

0.72

 tracing

0.70

 booths

0.69

oleon

0.67

eering

0.67

afety

0.66

Activations Density 0.009%

references to the concept of profiling, especially in the context of racial profiling and computer programming

terms related to profiling in various contexts

No Comments

No Known Activations