INDEX

Explanations

positive character traits and ethical values

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

nesium

-0.70

cise

-0.67

ergy

-0.63

ODE

-0.59

onga

-0.58

asel

-0.58

JECT

-0.57

ERSON

-0.56

Bed

-0.54

dule

-0.54

POSITIVE LOGITS

fulness

1.34

lessness

1.25

iness

1.15

worthiness

1.10

uality

1.01

iveness

0.98

 quot

0.97

FUL

0.97

ful

0.93

 perv

0.91

Activations Density 2.767%

positive character traits and ethical values

No Comments

No Known Activations