INDEX

Explanations

numbers within a range of 0 to 10 that represent important aspects of a situation or an object

oai_token-act-pair · gpt-3.5-turbo

repeated phrases and concepts associated with the word "another"

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

 Clubs

-0.67

 Ajax

-0.65

 Sporting

-0.62

 Moonlight

-0.62

 Reprodu

-0.62

ople

-0.61

 Apprentice

-0.60

 Hobby

-0.60

 Symbol

-0.58

 Reporter

-0.58

POSITIVE LOGITS

worldly

0.99

roller

0.85

anche

0.83

ioch

0.80

than

0.78

ypes

0.76

rolling

0.75

 than

0.75

 besides

0.73

blance

0.72

Activations Density 0.247%

numbers within a range of 0 to 10 that represent important aspects of a situation or an object

repeated phrases and concepts associated with the word "another"

No Comments

No Known Activations