INDEX

Explanations

mentions of specific decades of the past

oai_token-act-pair · gpt-3.5-turbo

the decade of the 1970s referenced multiple times in a document

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

edIn

-0.81

ournals

-0.72

 snipp

-0.69

rotein

-0.67

 parcels

-0.67

Maker

-0.65

phabet

-0.65

acts

-0.64

 clipboard

-0.63

imentary

-0.63

POSITIVE LOGITS

 onwards

1.06

 onward

1.05

ixties

0.87

era

0.83

Era

0.81

-'

0.71

 Barkley

0.71

 Patton

0.68

lander

0.66

 nostalg

0.63

Activations Density 0.033%

mentions of specific decades of the past

the decade of the 1970s referenced multiple times in a document

No Comments

No Known Activations