INDEX

Explanations

words related to examining or exploring topics

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "Nexus" related to various contexts or entities

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

jriwal

-0.80

riter

-0.79

OOD

-0.73

 Rohing

-0.64

¶ħ

-0.63

IRD

-0.63

anche

-0.62

 roofs

-0.61

Sut

-0.61

 Sawyer

-0.60

POSITIVE LOGITS

actly

1.23

posure

1.22

cellent

1.20

haust

1.10

ercise

1.09

change

1.08

odus

1.07

cellence

1.04

xon

1.01

clusively

1.00

Activations Density 0.014%

words related to examining or exploring topics

mentions of the term "Nexus" related to various contexts or entities

No Comments

No Known Activations