INDEX

Explanations

mentions of location such as cities or regions

oai_token-act-pair · gpt-3.5-turbo

references to time periods and locations

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

 someday

-0.74

isma

-0.63

ragon

-0.62

anium

-0.59

 unemploy

-0.59

emouth

-0.58

 somehow

-0.58

 infring

-0.57

Ore

-0.57

 choke

-0.57

POSITIVE LOGITS

 meanwhile

0.88

ebus

0.70

VIEW

0.68

 Scene

0.67

 meantime

0.66

enture

0.64

ournal

0.62

idav

0.60

 Inquis

0.60

 Gloss

0.59

Activations Density 0.322%

mentions of location such as cities or regions

references to time periods and locations

No Comments

No Known Activations