INDEX

Explanations

words related to a specific location, possibly a city or a soccer team

oai_token-act-pair · gpt-3.5-turbo

references to specific locations and terms associated with geography

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uve

-0.72

anche

-0.69

 McAuliffe

-0.66

vi

-0.66

 Bucc

-0.64

ress

-0.64

 Germ

-0.63

OPLE

-0.63

 Barron

-0.61

 Pruitt

-0.60

POSITIVE LOGITS

seys

1.43

ynthesis

1.17

hips

1.16

hops

1.11

mith

1.09

pring

1.05

creen

1.05

hip

1.02

ystem

0.97

peak

0.96

Activations Density 0.030%

words related to a specific location, possibly a city or a soccer team

references to specific locations and terms associated with geography

No Comments

No Known Activations