INDEX

Explanations

phrases related to "returning to" a place, team, or role

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a return to a previous state or place

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

merce

-0.77

holes

-0.71

VPN

-0.68

riad

-0.64

redict

-0.64

Secondly

-0.63

lied

-0.63

erity

-0.63

croft

-0.61

ially

-0.61

POSITIVE LOGITS

 haunt

1.07

 prominence

0.94

 earth

0.83

 shore

0.83

 obscurity

0.83

 sender

0.80

 normal

0.76

 consciousness

0.76

 fold

0.74

 fray

0.73

Activations Density 0.074%

phrases related to "returning to" a place, team, or role

phrases indicating a return to a previous state or place

No Comments

No Known Activations