INDEX

Explanations

short phrases starting with "to" that describe accomplishments or actions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating first occurrences or unique achievements

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

soDeliveryDate

-0.73

 ILCS

-0.69

ãĤ¦ãĤ¹

-0.67

 Definitions

-0.66

maps

-0.65

Apps

-0.62

CI

-0.62

tun

-0.62

 Continued

-0.62

course

-0.60

POSITIVE LOGITS

 arrive

1.26

 receive

1.24

 endorse

1.19

 emerge

1.18

 succumb

1.18

 undergo

1.15

 admit

1.11

 criticize

1.09

 suffer

1.09

 declare

1.08

Activations Density 0.083%

short phrases starting with "to" that describe accomplishments or actions

phrases indicating first occurrences or unique achievements

No Comments

No Known Activations