INDEX

Explanations

adjectives and adverbs related to levels, expenses, and positions

oai_token-act-pair · gpt-3.5-turbo

terms related to levels, density, and cost

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

afia

-0.72

idav

-0.64

 Goods

-0.63

 Guards

-0.60

andals

-0.60

 Seym

-0.60

 Built

-0.60

iren

-0.59

 Messenger

-0.58

 Invaders

-0.58

POSITIVE LOGITS

 intervals

1.32

 speeds

1.32

 expense

1.26

 levels

1.17

 pace

1.16

 rates

1.14

 distances

1.10

 heights

1.09

 altitude

1.08

 angles

1.05

Activations Density 0.139%

adjectives and adverbs related to levels, expenses, and positions

terms related to levels, density, and cost

No Comments

No Known Activations