INDEX

Explanations

place names and location-related terms

oai_token-act-pair · gpt-3.5-turbo

references to specific locations and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

DeliveryDate

-1.10

ortunately

-1.02

 teasp

-0.97

 exting

-0.97

 eleph

-0.96

aditional

-0.95

oreAnd

-0.91

ÃĥÃĤÃĥÃĤ

-0.90

 unnecess

-0.89

ailability

-0.88

POSITIVE LOGITS

od

0.71

op

0.71

open

0.67

ian

0.67

auder

0.66

 Arcane

0.66

ine

0.64

NEO

0.64

ic

0.64

ocry

0.60

Activations Density 0.185%

place names and location-related terms

references to specific locations and events

No Comments

No Known Activations