INDEX

Explanations

terms related to hard drives

oai_token-act-pair · gpt-3.5-turbo

references to hard drives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 McCull

-0.77

amera

-0.75

 Scha

-0.71

 Nest

-0.70

 Sach

-0.66

[|

-0.65

 Kern

-0.64

 Gould

-0.62

 Mara

-0.62

 Andromeda

-0.60

POSITIVE LOGITS

disk

1.22

copy

1.19

cover

1.18

wired

1.16

woods

1.14

dri

1.14

coded

1.13

drive

1.12

wood

1.12

core

1.03

Activations Density 0.032%

terms related to hard drives

references to hard drives

No Comments

No Known Activations