INDEX

Explanations

network-related terms and protocols

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

hips

-0.75

IGHTS

-0.74

âĸ¬âĸ¬

-0.69

 Lauder

-0.68

âĸ¬

-0.68

BIP

-0.67

beit

-0.64

 EntityItem

-0.64

taboola

-0.63

Â¯

-0.60

POSITIVE LOGITS

olitan

1.59

olitics

1.42

odcast

1.40

onent

1.39

ulse

1.39

osit

1.37

rint

1.36

ocalypse

1.36

olit

1.34

oleon

1.34

Activations Density 0.808%

network-related terms and protocols

No Comments

No Known Activations