INDEX

Explanations

keywords related to communication, news, and forwarding

oai_token-act-pair · gpt-3.5-turbo

phrases related to the concept of "forward" or "forwarding" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

chens

-0.73

arcity

-0.73

ALE

-0.73

avering

-0.71

redit

-0.68

earable

-0.68

gins

-0.66

uff

-0.65

adle

-0.64

ellen

-0.64

POSITIVE LOGITS

forward

1.34

 forwarding

1.00

 Forward

0.98

 forwarded

0.95

Forward

0.94

£ı

0.87

etheless

0.84

olicy

0.80

 forwards

0.78

folios

0.76

Activations Density 0.010%

keywords related to communication, news, and forwarding

phrases related to the concept of "forward" or "forwarding" in various contexts

No Comments

No Known Activations