INDEX

Explanations

phrases related to comparing, updating, or modifying existing things or concepts

oai_token-act-pair · gpt-3.5-turbo

references to current systems, regulations, or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

jah

-0.72

 salute

-0.66

 consolation

-0.63

zag

-0.60

oops

-0.59

zes

-0.58

nton

-0.57

 wonders

-0.57

Sort

-0.57

lus

-0.57

POSITIVE LOGITS

 existing

3.66

existing

2.45

 extant

1.64

 pree

1.54

 current

1.39

old

1.26

 prevailing

1.23

 remaining

1.18

 already

1.17

current

1.16

Activations Density 0.014%

phrases related to comparing, updating, or modifying existing things or concepts

references to current systems, regulations, or conditions

No Comments

No Known Activations