INDEX

Explanations

phrases related to sequels or subsequent events

oai_token-act-pair · gpt-3.5-turbo

references to follow-up actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 rusty

-0.68

 constitu

-0.66

zsche

-0.61

 Klux

-0.61

 privately

-0.60

 Cotton

-0.60

 DRAG

-0.60

 opener

-0.59

 Norris

-0.57

 soluble

-0.57

POSITIVE LOGITS

advertisement

1.03

release

0.99

up

0.95

appro

0.94

beat

0.86

action

0.86

through

0.86

growth

0.85

payment

0.83

ste

0.83

Activations Density 0.028%

phrases related to sequels or subsequent events

references to follow-up actions or events

No Comments

No Known Activations