INDEX

Explanations

possessive pronouns followed by a location or entity

oai_token-act-pair · gpt-3.5-turbo

possessive pronouns and references to belonging

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

wcsstore

-0.86

agu

-0.83

Magikarp

-0.81

çĶ°

-0.78

ecd

-0.77

artifacts

-0.76

ocument

-0.75

clair

-0.74

Interstitial

-0.71

Versions

-0.71

POSITIVE LOGITS

 outset

1.22

 equation

1.12

 brink

1.05

 confines

1.02

 shores

1.01

 pedest

0.94

 fray

0.89

 bloodstream

0.89

 premises

0.89

 wreckage

0.88

Activations Density 0.153%

possessive pronouns followed by a location or entity

possessive pronouns and references to belonging

No Comments

No Known Activations