INDEX

Explanations

phrases indicating possession or existence of objects in a specific place

oai_token-act-pair · gpt-3.5-turbo

the phrase "in" appearing 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uces

-0.70

idays

-0.69

uden

-0.64

enance

-0.62

isky

-0.60

 VIDEOS

-0.60

wolves

-0.59

EMBER

-0.56

 presided

-0.56

0000000000000000

-0.56

POSITIVE LOGITS

 circulation

1.30

activated

1.21

 possession

1.09

accessible

1.08

 storage

1.04

operative

1.04

ked

1.04

 existence

0.99

 situ

0.98

escap

0.95

Activations Density 0.182%

phrases indicating possession or existence of objects in a specific place

the phrase "in" appearing in various contexts

No Comments

No Known Activations