INDEX

Explanations

phrases related to various items being included in a list

oai_token-act-pair · gpt-3.5-turbo

instances of the word "include" and its variations to indicate lists or examples

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

 proble

-0.72

acia

-0.70

iliate

-0.70

aptic

-0.70

aline

-0.68

ometer

-0.67

orean

-0.65

iet

-0.64

 exting

-0.64

rait

-0.63

POSITIVE LOGITS

 prominently

0.69

ãĤ¯

0.67

:'

0.61

ãĥ¯

0.60

 nods

0.59

krit

0.59

ãģĤ

0.59

staking

0.58

ESCO

0.57

:#

0.57

Activations Density 0.049%

phrases related to various items being included in a list

instances of the word "include" and its variations to indicate lists or examples

No Comments

No Known Activations