INDEX

Explanations

words related to specific objects or products

oai_token-act-pair · gpt-3.5-turbo

plural nouns and their variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

âĸ¬

-0.68

 Niet

-0.67

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.67

QC

-0.63

ISON

-0.63

 Jihad

-0.63

 streng

-0.63

Qual

-0.62

IENT

-0.61

oÄŁ

-0.60

POSITIVE LOGITS

hip

1.24

hips

1.23

mith

1.14

hell

1.13

etting

1.11

hare

1.11

avers

1.07

poons

1.05

cale

1.05

igmatic

1.05

Activations Density 0.266%

words related to specific objects or products

plural nouns and their variants

No Comments

No Known Activations