INDEX

Explanations

phrases indicating a strong emphasis or focus on something

oai_token-act-pair · gpt-3.5-turbo

the word "that" 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 @ 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

ATURES

-0.93

²¾

-0.90

ourses

-0.89

izont

-0.87

atures

-0.84

å§«

-0.84

IDES

-0.83

AMS

-0.82

orians

-0.79

zos

-0.79

POSITIVE LOGITS

 pesky

1.30

guy

1.03

 kind

1.01

 sucker

1.01

 fateful

0.99

 particular

0.96

 sort

0.96

 bastard

0.95

 elusive

0.94

 damned

0.94

Activations Density 0.129%

phrases indicating a strong emphasis or focus on something

the word "that" in various contexts

No Comments

No Known Activations