INDEX

Explanations

phrases indicating a decrease or reduction in something

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

DD

-0.64

ãĤ±

-0.62

SPONSORED

-0.61

 assemb

-0.60

 synthes

-0.57

00000000

-0.56

osc

-0.56

 slic

-0.56

âĹ¼

-0.56

alys

-0.55

POSITIVE LOGITS

 FILE

0.82

thumbnails

0.70

 than

0.69

 guiIcon

0.68

 âĶľâĶĢâĶĢ

0.67

 IMAGES

0.65

maxwell

0.65

enz

0.64

 VIDE

0.61

inez

0.61

Activations Density 0.007%

phrases indicating a decrease or reduction in something

No Comments

No Known Activations