INDEX

Explanations

phrases related to recurring events that happen consistently within a year

oai_token-act-pair · gpt-3.5-turbo

recurring references to annual statistics or measurements

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

holes

-0.82

ocard

-0.72

ergus

-0.70

imir

-0.69

reprene

-0.68

yp

-0.67

yle

-0.66

hole

-0.64

elo

-0.63

arnaev

-0.62

POSITIVE LOGITS

 thereafter

1.04

 conclud

0.84

 consecut

0.77

 proport

0.75

rics

0.74

 annually

0.71

 contrace

0.71

ãĥ£

0.70

 unden

0.70

iod

0.69

Activations Density 0.012%

phrases related to recurring events that happen consistently within a year

recurring references to annual statistics or measurements

No Comments

No Known Activations