INDEX

Explanations

phrases related to specific objects or concepts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the" and related phrases that emphasize frequency or significant moments

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

Cho

-0.68

chy

-0.64

cart

-0.63

ÙĴ

-0.62

jee

-0.61

itia

-0.61

wagen

-0.60

gall

-0.60

een

-0.60

quartered

-0.60

POSITIVE LOGITS

 sake

1.79

 foreseeable

1.43

 purposes

1.32

 duration

1.16

 purpose

1.15

 entirety

1.09

um

1.03

 remainder

1.00

 benefit

0.98

 reasons

0.97

Activations Density 0.092%

phrases related to specific objects or concepts

occurrences of the word "the" and related phrases that emphasize frequency or significant moments

No Comments

No Known Activations