INDEX

Explanations

mentions of a specific year

oai_token-act-pair · gpt-3.5-turbo

references to the current year

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 encl

-0.71

 textual

-0.69

atem

-0.67

ocker

-0.64

 cumbers

-0.63

 contag

-0.63

 sympath

-0.63

omething

-0.62

ICLE

-0.62

 zipper

-0.62

POSITIVE LOGITS

long

1.02

ago

0.93

Ago

0.91

 olds

0.89

iversary

0.83

nings

0.83

ns

0.75

eve

0.75

glass

0.75

dream

0.74

Activations Density 0.111%

mentions of a specific year

references to the current year

No Comments

No Known Activations