INDEX

Explanations

indications of positioning and objects being in a specific place or location

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "of"

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

PDF

-0.66

LOD

-0.65

>(

-0.64

>>>>>>>>

-0.62

0000000000000000

-0.62

heet

-0.60

ealous

-0.60

 thereto

-0.60

REF

-0.59

ctory

-0.59

POSITIVE LOGITS

 another

0.81

 unsuspecting

0.74

 slain

0.72

rium

0.72

irlf

0.70

 scrimmage

0.67

an

0.66

 deceased

0.66

 whichever

0.64

Activations Density 0.248%

indications of positioning and objects being in a specific place or location

phrases that include the word "of"

No Comments

No Known Activations