INDEX

Explanations

phrases related to temporal references and their relationships

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

orable

-0.72

 Brav

-0.70

ĸļ

-0.69

 Skies

-0.67

 unequivocally

-0.66

 Parables

-0.66

 dearly

-0.60

 foremost

-0.57

 unfl

-0.56

SEE

-0.56

POSITIVE LOGITS

å£«

0.90

 margins

0.70

 bend

0.70

clock

0.69

 halfway

0.69

 holidays

0.68

 margin

0.68

 ripe

0.67

âĸĪâĸĪ

0.66

topic

0.65

Activations Density 0.045%

phrases related to temporal references and their relationships

No Comments

No Known Activations