INDEX

Explanations

numerical comparisons or statistical information

oai_token-act-pair · gpt-3.5-turbo

statistical comparisons and mentions of mortality or health-related rates

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

ĪĴ

-0.88

 Redemption

-0.68

ADRA

-0.65

claimer

-0.65

 offerings

-0.62

sou

-0.60

 sails

-0.59

 SOFTWARE

-0.59

 Wonderland

-0.58

inventoryQuantity

-0.56

POSITIVE LOGITS

 among

1.16

 across

1.10

among

1.03

 amongst

1.00

 between

1.00

 nationally

1.00

 nationwide

0.97

 observed

0.88

(%

0.87

between

0.86

Activations Density 0.370%

numerical comparisons or statistical information

statistical comparisons and mentions of mortality or health-related rates

No Comments

No Known Activations