INDEX

Explanations

mentions of official visits or trips to various locations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "to" and related phrases indicating travel or visits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

standing

-0.74

 ticking

-0.70

etooth

-0.69

anni

-0.68

ratulations

-0.66

iction

-0.65

acters

-0.65

paced

-0.65

rified

-0.65

 quickest

-0.62

POSITIVE LOGITS

 Disneyland

0.95

 meet

0.95

tnc

0.82

 Thailand

0.81

 Osaka

0.80

 Uganda

0.79

 visit

0.78

 Dover

0.78

 Mecca

0.78

 Rome

0.77

Activations Density 0.126%

mentions of official visits or trips to various locations

occurrences of the word "to" and related phrases indicating travel or visits

No Comments

No Known Activations