INDEX

Explanations

questions or expressions of uncertainty

oai_token-act-pair · gpt-3.5-turbo

phrases that pose questions about understanding or explanations

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

Laughs

-0.72

otti

-0.68

OGR

-0.60

rive

-0.60

enza

-0.59

iva

-0.59

©¶æ¥µ

-0.58

ONG

-0.58

ey

-0.57

horn

-0.57

POSITIVE LOGITS

why

1.81

 whether

1.61

WHY

1.59

how

1.49

why

1.42

 whence

1.28

 what

1.27

whether

1.21

HOW

1.14

 whereabouts

1.12

Activations Density 0.174%

questions or expressions of uncertainty

phrases that pose questions about understanding or explanations

No Comments

No Known Activations