INDEX

Explanations

mentions of the word "Gun" in the text

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

 Strawberry

-0.75

ç«

-0.71

 Archdemon

-0.69

 Hort

-0.67

 Dragons

-0.67

 Virtue

-0.64

 coli

-0.63

 Humph

-0.60

 Advis

-0.59

uthor

-0.59

POSITIVE LOGITS

powder

1.74

nery

1.32

smith

1.23

ner

1.20

ners

1.16

nar

1.14

ning

1.12

fight

1.11

boats

1.11

fighters

1.11

Activations Density 0.032%

mentions of the word "Gun" in the text

No Comments

No Known Activations