INDEX

Explanations

information related to trips or travels, especially those involving specific locations or activities

oai_token-act-pair · gpt-3.5-turbo

references to travel or journeys

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

initions

-0.73

 relational

-0.71

 Canary

-0.68

 fledgling

-0.68

MpServer

-0.68

 denomin

-0.66

ocker

-0.66

 thriving

-0.64

confidence

-0.63

 flourishing

-0.61

POSITIVE LOGITS

 trip

1.21

trip

1.17

 trips

1.13

 Trip

0.92

 traveled

0.80

olit

0.79

 itiner

0.76

 trek

0.75

ogue

0.74

 travels

0.74

Activations Density 0.014%

information related to trips or travels, especially those involving specific locations or activities

references to travel or journeys

No Comments

No Known Activations