INDEX

Explanations

words related to transportation and locations, potentially focusing on negative instances

oai_token-act-pair · gpt-3.5-turbo

negative phrases related to various forms of loss or deprivation

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

ulhu

-0.62

 Dickinson

-0.55

AVG

-0.52

 EntityItem

-0.52

 Burr

-0.51

 Norris

-0.50

 Richards

-0.49

 Meadows

-0.49

 Borders

-0.48

 Rica

-0.47

POSITIVE LOGITS

sized

1.03

based

0.95

level

0.92

shaped

0.88

themed

0.88

grade

0.86

eyed

0.86

related

0.85

derived

0.84

powered

0.84

Activations Density 0.319%

words related to transportation and locations, potentially focusing on negative instances

negative phrases related to various forms of loss or deprivation

No Comments

No Known Activations