INDEX

Explanations

words related to locks

oai_token-act-pair · gpt-3.5-turbo

phrases related to locking mechanisms or security measures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

issance

-0.86

LV

-0.76

schild

-0.74

abama

-0.73

ilater

-0.72

enegger

-0.67

aukee

-0.66

ciation

-0.65

ional

-0.65

ãĤ¡

-0.65

POSITIVE LOGITS

picking

1.21

heed

1.15

pick

1.08

creen

1.03

door

1.02

downs

0.94

lear

0.93

step

0.93

hold

0.91

 horns

0.90

Activations Density 0.038%

words related to locks

phrases related to locking mechanisms or security measures

No Comments

No Known Activations