INDEX

Explanations

phrases related to legality and privacy

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

bies

-0.88

 Finish

-0.67

iscons

-0.67

udeau

-0.67

ð

-0.66

ertodd

-0.66

igraph

-0.66

Gap

-0.65

othal

-0.65

esm

-0.64

POSITIVE LOGITS

 hereby

1.06

 indeed

1.04

nt

1.03

not

1.00

 presently

0.95

NOT

0.94

omorphic

0.93

 currently

0.93

 defin

0.89

 definitely

0.88

Activations Density 1.028%

phrases related to legality and privacy

No Comments

No Known Activations