INDEX

Explanations

terms related to quantification and measurement in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

HOME

-0.77

 Archdemon

-0.69

 alive

-0.67

 Mansion

-0.65

lihood

-0.65

 blades

-0.62

 pine

-0.60

 Legs

-0.59

 Spit

-0.59

 Castle

-0.57

POSITIVE LOGITS

itatively

1.61

ifiable

1.33

ified

1.22

itative

1.15

ifying

1.14

ifies

1.11

ifier

1.03

idian

1.03

ifiers

0.99

eros

0.98

Activations Density 0.005%

terms related to quantification and measurement in various contexts

No Comments

No Known Activations