INDEX

Explanations

phrases related to design elements

oai_token-act-pair · gpt-3.5-turbo

references to various design concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ieri

-0.69

 Sheen

-0.69

 Arist

-0.68

ishops

-0.67

Cah

-0.64

 Lauder

-0.63

essional

-0.63

lehem

-0.63

 Witnesses

-0.62

 ILCS

-0.61

POSITIVE LOGITS

 designs

0.90

heet

0.87

plates

0.84

paces

0.83

 layouts

0.82

hops

0.79

ators

0.78

isine

0.77

 etched

0.77

 tailored

0.76

Activations Density 0.022%

phrases related to design elements

references to various design concepts

No Comments

No Known Activations