INDEX

Explanations

references to groups of people or individuals

oai_token-act-pair · gpt-3.5-turbo

the frequent usage of the term "folks."

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

SER

-0.67

 Quran

-0.64

Pakistan

-0.64

series

-0.62

Published

-0.62

Course

-0.61

Stre

-0.61

Prom

-0.59

ESV

-0.58

Relations

-0.58

POSITIVE LOGITS

 folks

0.93

folk

0.86

rats

0.83

hip

0.77

who

0.76

kies

0.73

cape

0.73

wagon

0.72

ometimes

0.69

ourcing

0.69

Activations Density 0.013%

references to groups of people or individuals

the frequent usage of the term "folks."

No Comments

No Known Activations