INDEX

Explanations

phrases related to duration or length

oai_token-act-pair · gpt-3.5-turbo

references to duration or extended timeframes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Stars

-0.80

Ro

-0.79

elf

-0.78

Always

-0.74

ery

-0.73

IRO

-0.72

ILLE

-0.72

eers

-0.72

Sov

-0.70

mir

-0.70

POSITIVE LOGITS

 than

1.19

than

1.00

 Than

0.87

 longer

0.87

 periods

0.85

 stretches

0.83

 distances

0.82

 lasting

0.81

dated

0.79

 stint

0.78

Activations Density 0.010%

phrases related to duration or length

references to duration or extended timeframes

No Comments

No Known Activations