INDEX

Explanations

emotional or subjective words related to personal experiences

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NCT

-0.67

 Strike

-0.67

 Jackets

-0.66

 Generator

-0.65

 Decoder

-0.64

ibraries

-0.63

 Devils

-0.62

 Waves

-0.61

 Summit

-0.61

 Dragons

-0.60

POSITIVE LOGITS

urable

1.23

isable

1.20

istent

1.19

ishable

1.18

vable

1.18

ivable

1.17

ailable

1.17

ifiable

1.16

iable

1.15

ably

1.13

Activations Density 2.739%

emotional or subjective words related to personal experiences

No Comments

No Known Activations