INDEX

Explanations

verbs related to actions or behaviors

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

Ã¨re

-0.93

eria

-0.77

si

-0.71

see

-0.67

dt

-0.66

redo

-0.64

tes

-0.63

 requesting

-0.63

 covering

-0.62

eka

-0.62

POSITIVE LOGITS

 oneself

1.03

redients

0.89

BuyableInstoreAndOnline

0.80

 engagements

0.74

BUS

0.68

 Yourself

0.66

 yourself

0.63

 strangers

0.63

HAM

0.61

akespeare

0.59

Activations Density 2.049%

verbs related to actions or behaviors

No Comments

No Known Activations