INDEX

Explanations

words ending with 'o'

oai_token-act-pair · gpt-3.5-turbo

the letter "o" in a variety of contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 glim

-0.76

ĸļ

-0.75

IDENT

-0.69

 EntityItem

-0.69

 Shades

-0.69

 Parables

-0.68

è£

-0.67

ãĤ¼ãĤ¦ãĤ¹

-0.67

 compe

-0.66

 detrim

-0.66

POSITIVE LOGITS

zzi

1.32

zzo

1.27

ichi

1.06

onga

1.05

ghan

1.01

tto

1.01

zza

1.01

aks

1.00

cean

0.98

ope

0.98

Activations Density 0.053%

words ending with 'o'

the letter "o" in a variety of contexts

No Comments

No Known Activations