INDEX

Explanations

keywords related to locations or events

oai_token-act-pair · gpt-3.5-turbo

the verb "is" and its variations as indicators of state or existence

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

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

 Antar

-0.60

edIn

-0.58

weights

-0.56

 Autob

-0.55

 senses

-0.55

 Seym

-0.55

eor

-0.54

 IMAGES

-0.54

atern

-0.53

 Benn

-0.53

POSITIVE LOGITS

rael

1.29

omorphic

1.18

senal

1.04

ometric

1.03

lam

1.02

othermal

1.01

olation

1.01

abella

0.98

olate

0.96

tis

0.94

Activations Density 0.212%

keywords related to locations or events

the verb "is" and its variations as indicators of state or existence

No Comments

No Known Activations