INDEX

Explanations

references to specific historical time periods, particularly focusing on centuries

oai_token-act-pair · gpt-3.5-turbo

mentions of specific centuries in historical contexts

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

ECTION

-0.81

azor

-0.78

iquid

-0.75

ticket

-0.74

NYSE

-0.74

hod

-0.74

SourceFile

-0.72

Pack

-0.72

SpaceEngineers

-0.71

*/(

-0.70

POSITIVE LOGITS

 Revival

0.95

 onwards

0.85

BCE

0.81

 onward

0.81

era

0.79

Era

0.75

 Frankfurt

0.74

 Britons

0.74

 Romanian

0.73

 Galile

0.73

Activations Density 0.040%

references to specific historical time periods, particularly focusing on centuries

mentions of specific centuries in historical contexts

No Comments

No Known Activations