INDEX

Explanations

numbers and quantities mentioned in a document

oai_token-act-pair · gpt-3.5-turbo

numerical figures representing quantities, especially related to populations and statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

cdn

-0.66

 Another

-0.61

 Legislation

-0.59

thren

-0.59

 Programming

-0.59

 differs

-0.59

 Practices

-0.59

ontent

-0.59

yll

-0.58

format

-0.58

POSITIVE LOGITS

 ones

1.01

 aforementioned

0.76

oÄŁ

0.73

aband

0.66

apo

0.66

ikini

0.64

assad

0.64

 infamous

0.63

 aborted

0.62

enegger

0.61

Activations Density 0.282%

numbers and quantities mentioned in a document

numerical figures representing quantities, especially related to populations and statistics

No Comments

No Known Activations