INDEX

Explanations

names of specific individuals

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

ĺħ

-0.80

pell

-0.77

iful

-0.73

ļé

-0.71

¿½

-0.68

peed

-0.68

civ

-0.68

trak

-0.67

gom

-0.67

planes

-0.66

POSITIVE LOGITS

 McGregor

1.05

Smy

0.84

mire

0.83

thia

0.83

shaw

0.82

icut

0.79

 Connor

0.78

 McKenzie

0.77

olly

0.76

 Byrne

0.74

Activations Density 0.023%

names of specific individuals

No Comments

No Known Activations