INDEX

Explanations

names of political figures or terms related to political events

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

glers

-0.94

ahime

-0.75

istically

-0.66

[|

-0.66

ERY

-0.65

VILLE

-0.65

phal

-0.62

beit

-0.62

 Cage

-0.61

 Leap

-0.61

POSITIVE LOGITS

orters

1.50

rint

1.36

orter

1.30

rieve

1.28

utations

1.25

ublic

1.22

utation

1.21

ository

1.20

orted

1.16

atri

1.16

Activations Density 0.510%

names of political figures or terms related to political events

No Comments

No Known Activations