INDEX

Explanations

notes or comments within a text

oai_token-act-pair · gpt-3.5-turbo

references to notes or additional information

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

"},"

-0.67

aden

-0.66

 guiActiveUnfocused

-0.65

ãĤ¼ãĤ¦ãĤ¹

-0.63

ãĥı

-0.59

 hemor

-0.58

 plurality

-0.57

aced

-0.55

ãĤ°

-0.55

ãĥĩ

-0.54

POSITIVE LOGITS

!:

1.19

1.12

 Regarding

1.12

:-

1.09

*:

1.07

 disclaimer

1.06

.:

1.02

 note

0.97

 NOTE

0.95

 caveat

0.91

Activations Density 0.138%

notes or comments within a text

references to notes or additional information

No Comments

No Known Activations