INDEX

Explanations

places or locations

oai_token-act-pair · gpt-3.5-turbo

locations and venues associated with various events or activities

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

 fuss

-0.68

ratulations

-0.67

 pudding

-0.63

 formula

-0.62

hemy

-0.61

 breed

-0.61

 progress

-0.61

 traveller

-0.60

STEM

-0.60

ians

-0.60

POSITIVE LOGITS

 headquarters

0.94

itent

0.89

 Headquarters

0.83

 behest

0.82

ertodd

0.80

HQ

0.78

 helm

0.78

 Armory

0.76

ortmund

0.73

ĳå£«

0.72

Activations Density 0.515%

places or locations

locations and venues associated with various events or activities

No Comments

No Known Activations