INDEX

Explanations

names or terms related to locations

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

lde

-0.60

IPO

-0.59

itiveness

-0.57

 magnification

-0.56

glers

-0.56

eport

-0.55

luster

-0.55

 Seymour

-0.54

 veil

-0.54

elines

-0.54

POSITIVE LOGITS

azeera

1.21

arez

1.08

aeda

1.00

querque

0.86

hyde

0.85

aida

0.81

rity

0.81

usra

0.80

ighty

0.80

abet

0.75

Activations Density 0.082%

names or terms related to locations

No Comments

No Known Activations