INDEX

Explanations

large numbers in the context of quantities or statistics

oai_token-act-pair · gpt-3.5-turbo

references to quantities in millions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 distingu

-0.70

Discuss

-0.70

 Dating

-0.69

krit

-0.66

imen

-0.66

ourage

-0.65

agall

-0.64

ainer

-0.64

 explan

-0.63

bara

-0.63

POSITIVE LOGITS

 dollars

1.12

 hectares

1.06

 tonnes

0.98

 acres

0.97

aires

0.97

 pounds

0.97

sq

0.93

lbs

0.92

 gallons

0.92

 miles

0.88

Activations Density 0.063%

large numbers in the context of quantities or statistics

references to quantities in millions

No Comments

No Known Activations