INDEX

Explanations

personal pronouns and forms of the verb "to be" associated with self-identification

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

Zup

-0.73

çīĪ

-0.72

 Skydragon

-0.69

 Tanz

-0.68

 Objective

-0.66

 seiz

-0.64

Tek

-0.64

 fortun

-0.62

 sights

-0.62

ãĥ¼ãĥĨ

-0.62

POSITIVE LOGITS

Ļ

1.27

ª

1.17

¤

1.14

¬

1.14

£

1.12

ħ

1.10

Ķ

1.09

ĸ

1.09

¼

1.07

Ĵ

1.05

Activations Density 0.240%

personal pronouns and forms of the verb "to be" associated with self-identification

No Comments

No Known Activations