INDEX

Explanations

phrases related to abstract concepts

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate comprehension or knowledge

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

onies

-0.83

nar

-0.70

ads

-0.68

sites

-0.63

drops

-0.61

adding

-0.61

Textures

-0.61

 piling

-0.60

yss

-0.60

quer

-0.60

POSITIVE LOGITS

ually

0.94

 Understanding

0.86

 understanding

0.85

 comprehension

0.80

ably

0.80

 Understand

0.78

how

0.74

HOW

0.74

FUL

0.74

displayText

0.72

Activations Density 0.011%

phrases related to abstract concepts

phrases that indicate comprehension or knowledge

No Comments

No Known Activations