INDEX

Explanations

phrases related to works or functioning

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of the word "Works" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 limb

-0.75

 imply

-0.66

 taboo

-0.63

 intr

-0.62

jay

-0.62

ent

-0.62

 bone

-0.62

uca

-0.62

 chasing

-0.60

pr

-0.60

POSITIVE LOGITS

 Works

3.99

Works

2.40

works

2.34

 WORK

1.77

 works

1.72

 Work

1.62

WORK

1.51

 Plays

1.27

Work

1.26

 Writ

1.24

Activations Density 0.013%

phrases related to works or functioning

the repeated mention of the word "Works" in various contexts

No Comments

No Known Activations