INDEX

Explanations

statistical comparisons indicating likelihood or probability

oai_token-act-pair · gpt-3.5-turbo

comparative phrases related to demographic statistics and behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

enment

-0.70

ablishment

-0.68

 Refresh

-0.67

ionics

-0.67

emonium

-0.65

acly

-0.65

imentary

-0.64

anian

-0.64

 Wonderful

-0.63

seq

-0.62

POSITIVE LOGITS

 disproportionately

1.25

 happier

1.19

 happiest

1.11

 disproportion

1.09

 wealthier

1.02

 healthier

1.01

 significantly

0.99

 discriminated

0.98

 more

0.96

 twice

0.96

Activations Density 0.143%

statistical comparisons indicating likelihood or probability

comparative phrases related to demographic statistics and behaviors

No Comments

No Known Activations