INDEX

Explanations

references to footwear items

oai_token-act-pair · gpt-3.5-turbo

references to "boot" or related terms

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Newsp

-0.73

terday

-0.70

NRS

-0.70

 neurot

-0.69

NAD

-0.67

Ĭ±

-0.66

ittee

-0.66

NH

-0.65

LD

-0.64

Pok

-0.63

POSITIVE LOGITS

strap

1.61

loader

1.05

 boot

0.97

stra

0.90

boot

0.89

legged

0.88

Boot

0.81

camp

0.80

 loader

0.78

lake

0.78

Activations Density 0.004%

references to footwear items

references to "boot" or related terms

No Comments

No Known Activations