INDEX

Explanations

text related to experiences

oai_token-act-pair · gpt-3.5-turbo

instances of the word "experience" or its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 shade

-0.68

 seal

-0.67

tie

-0.66

 Windsor

-0.66

 Link

-0.65

 hangs

-0.65

 solicitation

-0.65

 Lawn

-0.63

 framed

-0.63

 nods

-0.62

POSITIVE LOGITS

exper

4.23

Exper

2.29

 Exper

1.95

 experien

1.41

Experience

1.38

 Experiment

1.27

 Experience

1.25

 exper

1.20

 experienced

1.14

 experiment

1.07

Activations Density 0.023%

text related to experiences

instances of the word "experience" or its variants

No Comments

No Known Activations