INDEX

Explanations

comparisons related to quantity or degree, specifically focusing on the concept of "less."

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a reduction in quantity or cost

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

DK

-0.72

 Events

-0.69

âĹ¼

-0.68

Draft

-0.66

 Advis

-0.63

 Properties

-0.62

omnia

-0.62

de

-0.62

Spons

-0.61

SU

-0.61

POSITIVE LOGITS

 than

1.12

ened

1.07

ening

0.96

 forgiving

0.76

 glamorous

0.70

than

0.68

 behaved

0.68

 detectable

0.67

 intrusive

0.66

 conspicuous

0.66

Activations Density 0.036%

comparisons related to quantity or degree, specifically focusing on the concept of "less."

phrases indicating a reduction in quantity or cost

No Comments

No Known Activations