INDEX

Explanations

key terms related to additions or enhancements

oai_token-act-pair · gpt-3.5-turbo

references to additional features or enhancements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ards

-0.73

NING

-0.72

ARE

-0.72

cies

-0.69

bane

-0.68

yah

-0.66

bird

-0.66

Bey

-0.65

 Zimmer

-0.65

Zel

-0.63

POSITIVE LOGITS

endum

1.03

 insult

0.96

itionally

0.92

added

0.88

 impetus

0.86

ictions

0.85

itional

0.85

 thereto

0.84

itious

0.79

 bonus

0.79

Activations Density 0.032%

key terms related to additions or enhancements

references to additional features or enhancements

No Comments

No Known Activations