INDEX

Explanations

words related to the term "loss", potentially focusing on financial or emotional loss

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

vation

-0.71

ãĥ£

-0.67

 Nile

-0.65

ths

-0.64

rative

-0.63

ters

-0.63

ric

-0.62

rities

-0.61

HCR

-0.60

ting

-0.60

POSITIVE LOGITS

 Whedon

1.14

essed

1.10

enger

1.07

ack

1.05

essing

1.05

aic

1.01

ums

1.00

acks

1.00

es

0.97

pec

0.96

Activations Density 0.079%

words related to the term "loss", potentially focusing on financial or emotional loss

No Comments

No Known Activations