INDEX

Explanations

terms related to computer chips or technology hardware

oai_token-act-pair · gpt-3.5-turbo

mentions of various types of chips in technology contexts

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

æĢ

-0.73

 Stranger

-0.67

Vaj

-0.64

 Sacrament

-0.64

 Scandinav

-0.64

 Territ

-0.63

lihood

-0.62

 Kamp

-0.61

 gearing

-0.61

 Bagg

-0.61

POSITIVE LOGITS

otle

1.45

washer

1.04

 chip

0.93

chip

0.91

 implanted

0.89

mun

0.89

 chips

0.87

intosh

0.82

board

0.81

Chip

0.81

Activations Density 0.014%

terms related to computer chips or technology hardware

mentions of various types of chips in technology contexts

No Comments

No Known Activations