INDEX

Explanations

phrases related to an individual person taking a particular action

oai_token-act-pair · gpt-3.5-turbo

instances of the word "will" indicating future actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

-0.73

Lear

-0.65

ourke

-0.64

 misunderstanding

-0.63

quickShipAvailable

-0.63

reality

-0.63

HQ

-0.60

Weak

-0.60

 calcul

-0.60

 abstraction

-0.60

POSITIVE LOGITS

be

1.19

 continue

1.10

 undoubtedly

1.01

 gladly

0.97

 doubtless

0.97

ows

0.96

 remain

0.94

 begin

0.93

 unveil

0.93

 join

0.93

Activations Density 0.200%

phrases related to an individual person taking a particular action

instances of the word "will" indicating future actions or events

No Comments

No Known Activations