INDEX

Explanations

requests or instructions in the text

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

pires

-0.76

lings

-0.71

é¾

-0.70

visor

-0.69

IUM

-0.69

arc

-0.69

cler

-0.69

laus

-0.66

MpServer

-0.65

bent

-0.65

POSITIVE LOGITS

 Ignore

1.00

 forgive

0.96

 beware

0.96

 note

0.95

 excuse

0.95

 enable

0.90

 ignore

0.90

 refrain

0.89

 advise

0.89

 disregard

0.88

Activations Density 0.433%

requests or instructions in the text

No Comments

No Known Activations