INDEX

Explanations

sentences that present a statement followed by an observation or explanation

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

ittees

-0.72

pees

-0.69

displayText

-0.69

vous

-0.68

oise

-0.68

ILA

-0.66

incial

-0.66

ĪĴ

-0.65

ãĥ¯ãĥ³

-0.64

ÑĤ

-0.64

POSITIVE LOGITS

"[

1.65

 "âĢ¦

1.55

 "...

1.38

"'

1.26

'[

1.18

:"

1.12

"(

1.10

0.97

:[

0.96

""

0.94

Activations Density 0.273%

sentences that present a statement followed by an observation or explanation

No Comments

No Known Activations