INDEX

Explanations

phrases related to actions taken or decisions made by individuals or organizations

oai_token-act-pair · gpt-3.5-turbo

phrases related to actions taken by entities or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

gery

-0.83

toggle

-0.69

Gun

-0.66

Fey

-0.66

LOT

-0.65

geries

-0.64

fter

-0.64

 Columb

-0.63

 spoiler

-0.63

 Designer

-0.62

POSITIVE LOGITS

unes

0.72

asca

0.70

rodu

0.69

 costs

0.68

 raining

0.66

emis

0.65

Ru

0.64

 subpoena

0.64

umo

0.64

 volunte

0.63

Activations Density 0.453%

phrases related to actions taken or decisions made by individuals or organizations

phrases related to actions taken by entities or organizations

No Comments

No Known Activations