INDEX

Explanations

specific terms related to technical documents and plans, particularly those called "blueprints."

oai_token-act-pair · gpt-3.5-turbo

terms related to plans or designs

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

BuyableInstoreAndOnline

-0.90

 Galile

-0.90

agan

-0.78

ruciating

-0.76

atan

-0.72

ann

-0.71

NRS

-0.71

annis

-0.71

enged

-0.70

angers

-0.70

POSITIVE LOGITS

prints

1.08

print

0.87

ertodd

0.83

 heartbeat

0.82

ingham

0.79

 etched

0.79

 Reloaded

0.78

 footprints

0.77

itiveness

0.77

 footsteps

0.76

Activations Density 0.035%

specific terms related to technical documents and plans, particularly those called "blueprints."

terms related to plans or designs

No Comments

No Known Activations