INDEX

Explanations

phrases related to increase or improvement

oai_token-act-pair · gpt-3.5-turbo

instances of the term "up" indicating increases or growth

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

>"

-0.71

>]

-0.70

][/

-0.68

"]=>

-0.66

]"

-0.64

====

-0.64

gee

-0.63

>[

-0.62

 Abyss

-0.62

[/

-0.61

POSITIVE LOGITS

grades

1.07

stairs

1.03

dates

0.99

graded

0.96

grading

0.94

dating

0.94

rights

0.90

river

0.84

olicy

0.84

skirts

0.84

Activations Density 0.133%

phrases related to increase or improvement

instances of the term "up" indicating increases or growth

No Comments

No Known Activations