INDEX

Explanations

proper nouns

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

 Samoa

-0.63

 Haram

-0.63

 briefs

-0.62

IST

-0.62

stakes

-0.61

ĨĴ

-0.60

 Hemp

-0.60

sag

-0.59

HAEL

-0.59

 Ø§ÙĦ

-0.59

POSITIVE LOGITS

unciation

1.23

owned

1.18

ovation

1.17

unci

1.13

ouncing

1.08

ault

1.04

aming

1.03

issance

0.99

ounced

0.96

ounce

0.94

Activations Density 0.027%

proper nouns

No Comments

No Known Activations