INDEX

Explanations

phrases related to politics and international affairs

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 teasp

-0.84

reci

-0.80

ãĤ¦ãĤ¹

-0.74

xual

-0.74

 behaviours

-0.73

ogeneity

-0.70

 conting

-0.70

aucuses

-0.69

 relate

-0.68

ptoms

-0.68

POSITIVE LOGITS

 University

1.07

 Carnegie

1.05

 Georgetown

0.99

 Univ

0.98

Nex

0.98

 Cornell

0.97

 Harvard

0.96

 Purdue

0.95

NYU

0.95

 UCLA

0.95

Activations Density 3.527%

phrases related to politics and international affairs

No Comments

No Known Activations