INDEX

Explanations

phrases related to openness, especially those including the term 'open-ended'

oai_token-act-pair · gpt-3.5-turbo

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

 Lumpur

-0.70

 Brach

-0.69

 Tsarnaev

-0.69

 Lowry

-0.67

Phi

-0.66

 Mons

-0.66

 Chocobo

-0.65

 Plum

-0.65

 Mania

-0.64

 Khan

-0.64

POSITIVE LOGITS

ended

1.20

source

1.10

minded

1.08

facing

1.04

eyed

1.03

circ

1.03

skinned

0.99

advertising

0.97

door

0.96

poll

0.96

Activations Density 0.034%

phrases related to openness, especially those including the term 'open-ended'

No Comments

No Known Activations