INDEX

Explanations

instances of human awareness or observation

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

pires

-0.70

erenn

-0.65

DonaldTrump

-0.59

 Subscribe

-0.53

 meets

-0.53

angelo

-0.53

âĦ¢:

-0.52

ctuary

-0.52

yip

-0.51

azeera

-0.51

POSITIVE LOGITS

sequently

0.98

 remained

0.87

RESULTS

0.85

 consisted

0.83

 hindsight

0.83

didn

0.82

 continued

0.80

 ended

0.79

was

0.78

Eventually

0.78

Activations Density 0.550%

instances of human awareness or observation

No Comments

No Known Activations