INDEX

Explanations

phrases indicating physical location or direction

oai_token-act-pair · gpt-3.5-turbo

instances of the word "on" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ptive

-0.74

INESS

-0.69

amount

-0.69

çĶ°

-0.68

-0.67

oland

-0.64

binding

-0.64

ocus

-0.64

odies

-0.63

orically

-0.63

POSITIVE LOGITS

 behalf

1.33

 occasion

1.29

screen

1.09

 Sundays

1.06

 steroids

1.05

 weekends

1.02

coming

0.99

 Saturdays

0.99

eday

0.98

etime

0.98

Activations Density 0.243%

phrases indicating physical location or direction

instances of the word "on" in various contexts

No Comments

No Known Activations