INDEX

Explanations

phrases related to technical concepts or specific terms used in various fields

oai_token-act-pair · gpt-3.5-turbo

terms related to technical, descriptive, and analytical concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ãĥĥãĥī

-0.77

 anyway

-0.71

rather

-0.67

Ĥ

-0.67

ãģĻ

-0.66

åĮ

-0.65

ĸ

-0.65

uliffe

-0.64

BUT

-0.64

Ĭ±

-0.64

POSITIVE LOGITS

 consists

1.40

 comprises

1.28

 begins

1.22

 celebrates

1.17

 contains

1.17

 utilizes

1.16

 revolves

1.14

 encompasses

1.13

 refers

1.11

 combines

1.11

Activations Density 0.534%

phrases related to technical concepts or specific terms used in various fields

terms related to technical, descriptive, and analytical concepts

No Comments

No Known Activations