INDEX

Explanations

phrases or sentences containing the structure 'by -', usually followed by a set of numbers or other words

oai_token-act-pair · gpt-3.5-turbo

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

 salts

-0.76

 Panic

-0.74

 Rohingya

-0.69

 Hung

-0.68

 relocated

-0.67

 Reich

-0.67

 smugglers

-0.66

 Grill

-0.66

 Nicotine

-0.66

 Silk

-0.66

POSITIVE LOGITS

prison

1.25

episode

1.25

distance

1.24

vict

1.23

event

1.23

quarter

1.21

character

1.21

dist

1.20

committee

1.20

step

1.19

Activations Density 0.052%

phrases or sentences containing the structure 'by -', usually followed by a set of numbers or other words

No Comments

No Known Activations