INDEX

Explanations

terms related to gatherings of people, especially with a focus on crowd control

oai_token-act-pair · gpt-3.5-turbo

terms related to large gatherings of people

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

tein

-0.76

Deg

-0.72

 CASE

-0.64

ces

-0.64

Adv

-0.61

ishops

-0.61

nesota

-0.61

omic

-0.60

 Provincial

-0.60

vell

-0.60

POSITIVE LOGITS

ourced

1.30

ourcing

1.18

funded

0.92

fund

0.83

atics

0.80

source

0.78

naires

0.78

 gathers

0.77

funding

0.77

sale

0.74

Activations Density 0.020%

terms related to gatherings of people, especially with a focus on crowd control

terms related to large gatherings of people

No Comments

No Known Activations