INDEX

Explanations

mentions of the city of Oakland

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Oakland."

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

mble

-0.87

orsche

-0.75

gey

-0.74

ittance

-0.74

pard

-0.71

osate

-0.70

 Beir

-0.70

htaking

-0.69

SHIP

-0.66

ivably

-0.66

POSITIVE LOGITS

 Raiders

1.02

 Tribune

0.90

 Coliseum

0.83

 County

0.82

 Heights

0.82

 Oakland

0.82

 Unified

0.81

 Hills

0.80

 Skydragon

0.80

 Athletics

0.78

Activations Density 0.009%

mentions of the city of Oakland

occurrences of the term "Oakland."

No Comments

No Known Activations