INDEX

Explanations

words related to including or adding something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "include" or its variations

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

acia

-0.87

stem

-0.76

CLASSIFIED

-0.71

ongyang

-0.66

²¾

-0.66

bell

-0.65

orbit

-0.65

app

-0.64

aver

-0.63

 exting

-0.63

POSITIVE LOGITS

 prominently

0.80

 Include

0.78

clus

0.68

ively

0.68

 disclaim

0.66

ãĤ½

0.66

 Kinect

0.65

ãĤ´

0.64

ãĥ¯

0.64

 elements

0.63

Activations Density 0.060%

words related to including or adding something

instances of the word "include" or its variations

No Comments

No Known Activations