INDEX

Explanations

words ending with the suffix "-at"

oai_token-act-pair · gpt-3.5-turbo

a repeated focus on the token "at" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

SPONSORED

-0.69

 velvet

-0.65

 Tsukuyomi

-0.64

Winged

-0.63

 Leilan

-0.62

pmwiki

-0.61

 forks

-0.60

ãģ®éŃĶ

-0.59

PE

-0.59

å£«

-0.59

POSITIVE LOGITS

mosp

1.17

abase

1.06

oday

0.95

istical

0.95

ravel

0.95

arian

0.94

raf

0.94

apult

0.92

rix

0.92

ainment

0.91

Activations Density 0.026%

words ending with the suffix "-at"

a repeated focus on the token "at" in various contexts

No Comments

No Known Activations