INDEX

Explanations

details related to legal or regulatory issues as well as specific naming instances

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

grown

-0.63

Pinterest

-0.62

Russ

-0.62

isSpecialOrderable

-0.57

animate

-0.56

©

-0.56

Spread

-0.56

Availability

-0.55

 rehearsal

-0.55

 sear

-0.55

POSITIVE LOGITS

 told

0.97

 tells

0.95

 urged

0.95

 wrote

0.94

 commented

0.90

 testified

0.89

 says

0.87

told

0.86

 explains

0.86

 warns

0.85

Activations Density 0.111%

details related to legal or regulatory issues as well as specific naming instances

No Comments

No Known Activations