INDEX

Explanations

terms related to challenges or difficulty

oai_token-act-pair · gpt-3.5-turbo

references to the concept of difficulty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

rum

-0.77

allery

-0.73

eer

-0.73

 Brus

-0.70

rophe

-0.69

eur

-0.68

zsche

-0.68

ta

-0.67

ournal

-0.67

rone

-0.66

POSITIVE LOGITS

iculty

1.12

icult

0.97

 Flavoring

0.83

 adjustment

0.76

coded

0.75

olving

0.75

 Modes

0.74

hooting

0.73

ioned

0.72

 comprom

0.72

Activations Density 0.069%

terms related to challenges or difficulty

references to the concept of difficulty

No Comments

No Known Activations