INDEX

Explanations

words related to physical displacement or removal

oai_token-act-pair · gpt-3.5-turbo

terms related to displacement and despotism

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

anwhile

-0.92

 Gutenberg

-0.74

ramid

-0.71

ebook

-0.69

soDeliveryDate

-0.65

 Leap

-0.64

 McCann

-0.64

 Tens

-0.62

manship

-0.62

tone

-0.62

POSITIVE LOGITS

oved

0.90

 disl

0.88

ixed

0.85

ocated

0.84

ighter

0.79

ief

0.76

odge

0.75

º

0.74

®

0.74

ikes

0.74

Activations Density 0.017%

words related to physical displacement or removal

terms related to displacement and despotism

No Comments

No Known Activations