INDEX

Explanations

specific geographical locations

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate location

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

HTML

-0.82

alities

-0.65

 persists

-0.65

 promotions

-0.63

 PRODUCT

-0.62

podcast

-0.61

 affirmative

-0.61

 hiring

-0.61

 packaged

-0.61

Professional

-0.60

POSITIVE LOGITS

rium

1.13

las

1.05

 dusk

1.02

 least

0.97

 sunset

0.91

olls

0.90

oll

0.89

 Mile

0.87

hens

0.87

mosp

0.86

Activations Density 0.118%

specific geographical locations

phrases that indicate location

No Comments

No Known Activations