INDEX

Explanations

statements related to events or occurrences

oai_token-act-pair · gpt-3.5-turbo

instances of the verb "was" to indicate 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 @ 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

é¾įå

-0.70

 equations

-0.70

 senses

-0.67

 antibodies

-0.65

isms

-0.65

snap

-0.65

eny

-0.64

conn

-0.63

dylib

-0.62

 weaknesses

-0.61

POSITIVE LOGITS

 postponed

1.45

 hosted

1.42

 scheduled

1.34

 cancelled

1.33

 convened

1.33

 canceled

1.31

 attended

1.28

 conducted

1.25

 billed

1.24

 underway

1.23

Activations Density 0.154%

statements related to events or occurrences

instances of the verb "was" to indicate actions or events

No Comments

No Known Activations