INDEX

Explanations

items or products within a list

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

otle

-0.84

gerald

-0.78

enegger

-0.76

ured

-0.73

eful

-0.72

uration

-0.71

aukee

-0.70

urated

-0.69

isconsin

-0.69

uity

-0.68

POSITIVE LOGITS

nd

1.88

 thirds

1.33

ND

1.01

 halves

1.00

 externalToEVAOnly

0.81

0.79

0.77

aries

0.76

entary

0.72

Activations Density 4.254%

items or products within a list

No Comments

No Known Activations