INDEX

Explanations

the word "simple" and its variants

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

simple and straightforward language

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

vance

-0.79

 largeDownload

-0.75

hovah

-0.75

 extensively

-0.74

reon

-0.70

ibal

-0.69

ashington

-0.69

haw

-0.69

 vigorously

-0.68

raints

-0.67

POSITIVE LOGITS

tons

1.42

minded

0.98

ton

0.97

 minded

0.97

 arithmetic

0.94

 syrup

0.93

 explanation

0.84

wallet

0.82

ified

0.81

 elegance

0.80

Activations Density 0.044%

the word "simple" and its variants

simple and straightforward language

No Comments

No Known Activations