INDEX

Explanations

terms related to technical or scientific jargon, particularly ones starting with special characters

oai_token-act-pair · gpt-3.5-turbo

references to a specific character or entity, denoted by a special symbol

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

 Reprodu

-0.70

 Drawn

-0.68

 beck

-0.63

 suite

-0.63

Ket

-0.62

 detached

-0.62

 Elys

-0.62

 Manz

-0.62

 numbered

-0.61

foc

-0.60

POSITIVE LOGITS

ª

1.34

Ĳ

1.33

«

1.26

ĳ

1.20

Ĵ

1.20

ı

1.17

ľ

1.16

ł

1.11

Ķ

1.09

¤

1.06

Activations Density 0.156%

terms related to technical or scientific jargon, particularly ones starting with special characters

references to a specific character or entity, denoted by a special symbol

No Comments

No Known Activations