INDEX

Explanations

proper nouns related to locations, organizations, and individuals

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and terms associated with them

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Gi

-0.70

wrapper

-0.64

 garg

-0.64

Gi

-0.63

ãĤĬ

-0.63

¶ħ

-0.62

 Watts

-0.62

Tab

-0.61

 filtered

-0.60

Ky

-0.60

POSITIVE LOGITS

endon

4.22

oxic

1.57

iens

1.54

endez

1.37

elong

1.28

igue

1.13

oxicity

1.12

vae

1.07

ibia

1.05

oxin

1.00

Activations Density 0.040%

proper nouns related to locations, organizations, and individuals

references to specific individuals and terms associated with them

No Comments

No Known Activations