INDEX

Explanations

words ending in '-ing'

oai_token-act-pair · gpt-3.5-turbo

gerunds or present participles of verbs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĤ£

-0.81

 Weston

-0.71

ivari

-0.69

dayName

-0.69

 Lockheed

-0.68

imet

-0.68

kamp

-0.67

imester

-0.66

ãĥ¼ãĥ«

-0.66

ichita

-0.66

POSITIVE LOGITS

LY

1.28

GGGGGGGG

1.19

ING

1.19

NESS

1.18

HAM

1.17

ERAL

1.10

ER

0.99

IAL

0.99

INGS

0.98

LES

0.98

Activations Density 0.012%

words ending in '-ing'

gerunds or present participles of verbs

No Comments

No Known Activations