INDEX

Explanations

references to the country of India

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Indian."

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

-0.89

umbnails

-0.85

olicy

-0.83

achine

-0.80

ijn

-0.80

uer

-0.77

escription

-0.77

isters

-0.76

bley

-0.76

oppable

-0.76

POSITIVE LOGITS

apolis

1.24

 Ocean

0.94

ru

0.86

 tribes

0.82

 Affairs

0.81

 Tribe

0.80

 Nations

0.78

istani

0.77

 Kashmir

0.75

 Airlines

0.74

Activations Density 0.017%

references to the country of India

occurrences of the term "Indian."

No Comments

No Known Activations