INDEX

Explanations

phrases indicating duration or time spent in a particular context

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

sonian

-0.72

hao

-0.72

aan

-0.71

dash

-0.69

 Accessed

-0.68

ãĤµ

-0.67

å£

-0.64

clear

-0.63

 frust

-0.61

ãĥ

-0.61

POSITIVE LOGITS

 awhile

1.05

 decades

0.94

 years

0.89

 months

0.80

 centuries

0.80

 several

0.78

 stint

0.78

 millennia

0.77

 instance

0.77

 example

0.76

Activations Density 0.089%

phrases indicating duration or time spent in a particular context

No Comments

No Known Activations