INDEX

Explanations

phrases related to aerial activities or locations

oai_token-act-pair · gpt-3.5-turbo

references to geographical locations or movements in the airspace

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

ovan

-0.77

orah

-0.71

 diaper

-0.71

urger

-0.69

abis

-0.69

zel

-0.69

CHO

-0.68

Percent

-0.67

 inherit

-0.67

wcsstore

-0.66

POSITIVE LOGITS

 Siberia

0.93

 Yosemite

0.90

 Antarctica

0.90

planes

0.87

 Cornwall

0.86

 Yellowstone

0.85

shore

0.84

 radar

0.82

 landfall

0.82

 Bermuda

0.81

Activations Density 0.226%

phrases related to aerial activities or locations

references to geographical locations or movements in the airspace

No Comments

No Known Activations