INDEX

Explanations

occurrences of the word 'Ren' at varying activation levels

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Ren."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

stakes

-0.83

milo

-0.75

HAEL

-0.73

 Antar

-0.70

Ö¼

-0.69

 Samoa

-0.67

 Ø§ÙĦ

-0.67

£ı

-0.66

é¾įå¥ĳå£«

-0.66

OOL

-0.65

POSITIVE LOGITS

issance

1.14

unciation

1.08

unci

1.02

ovation

0.99

ouncing

0.93

ault

0.90

ivers

0.88

emy

0.87

aming

0.87

semble

0.86

Activations Density 0.010%

occurrences of the word 'Ren' at varying activation levels

mentions of the name "Ren."

No Comments

No Known Activations